Research Article  Open Access
Egill Thorbergsson, Tomas Grönstedt, "A Thermodynamic Analysis of Two Competing MidSized Oxyfuel Combustion Combined Cycles", Journal of Energy, vol. 2016, Article ID 2438431, 14 pages, 2016. https://doi.org/10.1155/2016/2438431
A Thermodynamic Analysis of Two Competing MidSized Oxyfuel Combustion Combined Cycles
Abstract
A comparative analysis of two midsized oxyfuel combustion combined cycles is performed. The two cycles are the semiclosed oxyfuel combustion combined cycle (SCOCCC) and the Graz cycle. In addition, a reference cycle was established as the basis for the analysis of the oxyfuel combustion cycles. A parametric study was conducted where the pressure ratio and the turbine entry temperature were varied. The layout and the design of the SCOCCC are considerably simpler than the Graz cycle while it achieves the same net efficiency as the Graz cycle. The fact that the efficiencies for the two cycles are close to identical differs from previously reported work. Earlier studies have reported around a 3% points advantage in efficiency for the Graz cycle, which is attributed to the use of a second bottoming cycle. This additional feature is omitted to make the two cycles more comparable in terms of complexity. The Graz cycle has substantially lower pressure ratio at the optimum efficiency and has much higher power density for the gas turbine than both the reference cycle and the SCOCCC.
1. Introduction
The evidence that anthropogenically generated greenhouse gases are causing climate change is everincreasing. The Intergovernmental Panel on Climate Change (IPCC) has stated that [1]:
“It is extremely likely that more than half of the observed increase in global average surface temperature from 1951 to 2010 was caused by the anthropogenic increase in greenhouse gas concentrations and other types of anthropogenic forcing together. The best estimate of the human induced contribution to warming is similar to the observed warming over this period.”
One of the largest point source emitters of greenhouse gases is fossil fuel based power plants. One of the options to mitigate these greenhouse gases is to utilize Carbon Capture and Storage (CCS) in power plants.
There are three main processes being considered for CCS: the post combustion capture, the precombustion capture, and the oxyfuel combustion capture [2]. This paper focuses on the oxyfuel combustion combined cycle. The oxyfuel combustion fires fuel with nearly pure O_{2} instead of air and the resulting combustion products are primarily steam and carbon dioxide. This makes it technically more feasible to implement CO_{2} capturing solutions.
Two competing oxyfuel combustion combined cycles have shown promising potential [3]. These are the semiclosed oxyfuel combustion combined cycle (SCOCCC) and the Graz cycle. Numerous studies have been done on both of these cycles, including a number of studies that have compared the performance of the two cycles [3–37].
1.1. SCOCCC
The SCOCCC is essentially a combined cycle that uses nearly pure O_{2} as an oxidizer. After the heat recovery steam generator (HRSG) there is a condenser that condenses the water from the flue gas. The flue gas leaving the condenser is then primarily composed of CO_{2}. Part of the CO_{2} is then recycled back to the compressor while the rest is compressed and transported to a storage site.
Bolland and Sæther first introduced the SCOCCC concept in [4] where they compared new concepts for recovering CO_{2} from natural gas fired power plants. They compared a standard combined cycle with a cycle using postcombustion both with and without exhaust gas recirculation and also the SCOCCC, along with a Rankine cycle that incorporates oxyfuel combustion. Ulizar and Pilidis [5] were first to present a paper that focused exclusively on the performance of the SCOCCC. They started with cycle optimization and also simulated offdesign performance. They did more extensive work exploring the selection of an optimal cycle pressure ratio and turbine inlet temperature [6] and on operational aspects of the cycle [7]. Bolland and Mathieu also published studies on the SCOCCC concept [8] comparing its merits with a postcombustion removal plant. Amann et al. [9] also compared the SCOCCC with a combined cycle using a postcombustion plant and made a sensitivity analysis regarding the purity of the O_{2} and the corresponding energy cost of the air separation unit. Tak et al. compared the SCOCCC with a cycle developed by Clean Energy Systems and concluded that the SCOCCC seemed to be advantageous [10]. Jordal et al. proceeded to develop improved cooling flow prediction models [11] and Ulfsnes et al. studied transient operation [12] and further explored real gas effects and property modelling [13]. Other researchers have started with a conventional natural gas combined cycle as a starting point in the modelling of the SCOCCC. Riethmann et al. investigated the SCOCCC using a natural gas combined cycle as a reference case and concluded that the net efficiency of the SCOCCC was 8.3% points lower compared to the reference cycle [15]. Corchero et al. did a parametric study with regard to the pressure ratio at a fixed turbine entry temperature of 1327°C [14]. Yang et al. [16] modelled the SOCCCC along with the ASU and the CO_{2} compression train at different pressure ratios and two different turbine inlet temperatures, 1200°C and 1600°C. They concluded that the optimal pressure ratio is around 60 and 90 for the turbine inlet temperatures of 1418°C and 1600°C, respectively. With optimal design conditions, the net cycle efficiency is lower than the efficiency of the conventional CC by about 8 percentage points for both of the two turbine inlet temperatures. Dahlquist et al. optimized a midsized SCOCCC [17]. They concluded that although the optimum pressure ratio was 45 with regard to net efficiency, it would be beneficial to choose a lower pressure ratio for the cycle. Choosing a lower pressure ratio would only penalize the efficiency by a small amount but facilitate the design of the compressor by a great deal. Sammak et al. looked at different conceptual designs for the gas turbine for the SCOCCC [18]. They compared a single and twinshaft design and concluded that a twinshaft would be an advantageous design for the SCOCCC because of the high pressure ratio.
The main results of these papers have been summed together in Table 1. It is clear that there is no consensus in the power requirement of the air separation unit (ASU) as it ranges from 735 to 1440 kW/(kg/s) based on the mass flow of the O_{2} generated. This can be explained partly by the fact that the purity of the O_{2} stream varies from 90% to 96% and also that compression of the O_{2} stream is included in the ASU energy demand in some of the studies. The pressure ratio (PR) varies considerably between studies, where the lowest is 24.5 while the highest is 90. This can partly explain that only some of the studies aimed at optimizing the cycle with respect to pressure ratio, while others only selected a pressure ratio based on, for example, experience. Corchero et al. [14] and Yang et al. [16] looked at different pressure ratios and show similar trends as has been found in the current study. The resulting efficiencies from these studies have a very large spread and ranges from 36.7% to 53.9%. An important fact to take into consideration is that the oldest reference [4] is from 1992 and the stateoftheart cycle efficiency for a combined cycle at that time was much lower than is today. Bolland and Sæther found that the combined cycle efficiency, without carbon capture, was around 52% and that the SCOCCC was around 10% lower than the reference cycle. The studies that are presented in Table 1 are not based on the same assumptions, such as condenser cooling methods and turbine cooling which of course influence the net efficiencies reported.
 
Delivery pressure from ASU is the operating pressure. Delivery pressure from ASU is 1 bar. Delivery pressure from ASU is 27 bar. Delivery pressure from ASU is 1.2 bar. 
1.2. Graz Cycle
The basic principle of the Graz cycle was developed by Jericha in 1985 [19] and was aimed at solar generated O_{2}hydrogen fuel. Jericha et al. modified the cycle in 1995 [20] to handle fossil fuels.
The Graz cycle, similar to the SCOCCC, uses nearly pure O_{2} as oxidizer. In the Graz cycle a major part of the flue gas is recycled back to the compressor; that is, the water is not condensed out of it. Furthermore, the turbine and the combustion chamber are cooled using steam from the steam cycle. This means that the working fluid is mainly steam in the gas turbine. The other part of the flue gas goes through a condenser where the water is condensed from the flue gas. The flue gas is then in major parts CO_{2} which can be compressed and transported to a storage site. The full design of the Graz cycle incorporates a second bottoming cycle that uses the heat from the condensation of the flue gas. The second bottoming cycle is a subatmospheric steam cycle, since the condensation returns low quality heat, that is, low temperature.
There has been extensive research at the Graz University of Technology in designing the cycle, but the main focus has been on design of the turbomachinery components [21–29]. These studies have not been focused on optimizing the cycle performance with respect to pressure ratio. The main results of papers that study the Graz cycle are shown in Table 2. The design of the Graz cycle has been evolving and the most advanced cycle layout is the SGraz cycle which was presented in paper [26]. Publications published later all study the SGraz cycle concept.
 
Delivery pressure from ASU is the operating pressure. Delivery pressure from ASU is 1 bar. Energy cost for CO_{2} compression not taken into account. 
1.3. Comparison of Cycles
A number of papers have compared the two different cycles along with other carbon capture technologies for natural gas fired power plants [30–36]. The main results of papers that present cycle results are shown in Tables 1 and 2. Kvamsdal et al. compared nine different carbon capture options for natural gas fired power plants [30, 32]. Among them were the SCOCCC and the Graz cycle; the results for the cycle simulations are shown in Tables 1 and 2. It was concluded that concepts that employed very advanced technologies that have a low technological readiness level and high complexity achieved the highest performance. Franco et al. evaluated the technology feasibility of the components in 18 different novel power cycles with CO_{2} capture [31]. One of the conclusions was that the SCOCCC would be one of the cycles that incorporates gas turbines that would require the least effort to turn into a real power plant. Sanz et al. made a qualitative and quantitative comparison of the SCOCCC and the Graz cycle [33]. Their thermodynamic analysis showed that the hightemperature turbine of the SCOCCC plant needed a much higher cooling flow supply due to the less favourable properties of the working fluid than the Graz cycle turbine. They, in comparison to Franco et al. [31], concluded that all turbomachines of both cycles showed similar technical challenges and that the compressors and hightemperature turbines relied on new designs. Woollatt and Franco did a preliminary design study for both the compressor and the turbine, in both the SCOCCC and the Graz cycle [34]. They concluded that the turbomachinery can be designed using conventional levels of Mach number, hub/tip ratio, reaction, and flow and loading coefficients. They furthermore concluded that the efficiencies and the compressor surge margins of the components should be similar to a conventional gas turbine. Thorbergsson et al. examined both the Graz cycle and the SCOCCC [35]. They conceptually designed the compressor and the turbine for both cycles. They concluded that the Graz cycle, in the original version including the second bottoming cycles, is expected to be able to deliver around 3% points’ net efficiency benefit over the semiclosed oxyfuel combustion combined cycle at the expense of a more complex realization of the cycle.
Comparative work on the two cycles has suffered from not having the same technology level in the design of the two cycles. This results in the fact that it is difficult to draw conclusions from the comparisons. The aim of the current study is to assess the two cycles using the same technology level and in addition have comparable complexity levels. The current study goes into more details regarding the optimal pressure ratio and turbine entry temperatures for the oxyfuel combustion cycles then past publications. This is accomplished by establishing a reference cycle, which has a technology level that could enter service around year 2025. The fuel is assumed to be natural gas for all three cycles. The reference cycle is then used as the starting point for the modelling of the oxyfuel combustion combined cycles. In the current study the pressure ratio has been varied to locate the optimal net efficiency with respect to pressure ratio. Previous work has reported around 3% points’ benefit for the Graz cycle [32] including work carried out by the authors [35]. It was viewed that a majority of these benefits would be attributed to the use of a second bottoming cycle as included in the original implementation. To make a fair comparison of the two alternatives it was decided to exclude this cycle feature from the original Graz cycle. It should be noted that it is quite feasible to introduce such a bottoming concept also for the SCOCCC if the target would be to achieve maximum efficiency. The two simpler implementations were preferred in order to keep down complexity and make practical implementation more feasible.
2. Methods
The heat and mass balance program IPSEpro is used to simulate the power cycles [38]. The systems of equations, which are established using a graphical interface, are solved using a NewtonRaphson based algorithm. The simulation program was modified to incorporate the thermodynamic and transport properties program REFPROP to calculate the physical properties of fluids [39].
2.1. Cooled Turbine
2.1.1. Cooling Model
The cooling model is very important when studying the performance of gas turbine based cycles. The cooling model used is the model and is based on the work of Halls [40] and Holland and Thake [41]. The model is based on the standard blade assumption, which assumes that the blade has infinite thermal conductivity and a uniform blade temperature. The model used in this study was originally implemented by Jordal [42].
The main parameters for the cooling model are first the cooling efficiencywhere is the temperature of the cooling flow at the inlet, is the temperature of the cooling flow at the exit, and is the uniform blade temperature. The cooling efficiency is set to a moderate limit of .
Second the cooling effectiveness is defined aswhere is the hot gas temperature.
The model is a firstlaw thermodynamic, nondimensional model. The model is based on the dimensionless mass flow coolingwhere is the cooling mass flow, is the heat capacity of the cooling fluid, is the convective heat transfer coefficient on the hot gas side, and is the area of the blade. The main parameter of interest is the coolant mass flow ratiowhere is the heat capacity of the hot gas, is the average Stanton number of the hot gas, and is the crosssectional area of the hot gas. The relations between the cooling mass flow and the temperature differences areThe Stanton number is defined aswhere is the density of the hot gas and is the flow velocity.
To estimate the cooling requirements for each cooled turbine blade row, it was assumed that the cooling parameters were constant. The parameters were chosen to represent a cooled turbine that will enter service around 2025.
The uniform blade metal temperature is set to 850°C. This means that the maximum temperature will be around 950°C and the average temperature at the gas side of the blade around 900°C. The uniform blade metal temperature is used as the temperature limit for the cycle simulations.
It is assumed, as has been done in other studies [11, 43–46], that the Stanton number is constant, , in regard to both the change in the working fluid and the change in the design parameters. The parameters that are assumed to be constant in the cooling model are shown in Table 3. The geometry parameter, , which is the ratio between the wetted blade and adjacent cooled surface areas over the average gas crosssectional area, is also held constant between all cases. This parameter is unknown for a thermodynamic analysis where the key dimensions of the turbine have not been designed. ElMasri [44] estimated that this parameter is slightly less than 4.0 for a cascade blade row and around 8.0 for a stage, allowing for a rowtorow spacing. Jordal [47] concluded that when taking into account that rotor disks and the transition piece from the combustion chamber to the first stage nozzles are also subject to cooling, an average value should be around 5.0 for a stage.

The cooling model was used to reproduce the results in [48] and showed good agreement.
2.1.2. Expansion
The expansion in an uncooled turbine is modelled aswhere is the gas constant for the working fluid, , are the pressure and entropy, respectively, is the inlet, and is the outlet of the turbine stage. This model was evaluated against different models such as Mallen and Saville [49], using numerical integration and the model used gave good agreement with the numerical integration.
For the cooled turbine, the mixing of the coolant and the main stream gas flow result in a loss in stagnation pressure. This irreversibility is taken into account by defining a new polytropic efficiency [42, 50], defined aswhere is the stagnation pressure at the inlet of the rotor blade row, in is the inlet to the turbine, and out is the outlet of the turbine. Parameter is specific to each turbine and models the losses. It is typically in the range of for a turbine that has good performance and around for a turbine that has poor performance [51]. The polytropic efficiency is set to . The losses are taken into account by assuming that the factor is for all cases. Dahlquist et al. examined the empirical loss models used to design turbomachinery, which are generated using air as the working fluid, and concluded that the loss models generate similar results for the working fluids in oxyfuel cycles [52]. This indicates that it is possible to achieve a similar technology level for the oxyfuel turbines as for stateoftheart conventional turbines.
2.2. Compressor
The compression is modelled using polytropic efficiency,where is the gas constant for the working fluid, and are the pressure and entropy, respectively, is the inlet, and is the outlet of the compressor.
It is assumed that the polytropic efficiency is constant for all cycles and all cases and is assumed to be . Similar to the turbine, it is assumed that it is possible to achieve a compressor design for the oxyfuel compressor that is on the same level as the state of the art of compressors in conventional gas turbines.
2.3. Combustor
The combustion is a simple energy model based on the assumption that all of the fuel is completed in the combustion, that is, 100% combustion efficiency.
The amount of excess O_{2} is calculated aswhere is stoichiometric combustion. For the oxyfuel cycles the combustion is nearly stoichiometric; that is . It is preferred that the combustion takes place as close to stoichiometric conditions as possible to reduce the amount of O_{2} that the ASU needs to produce. Such a low amount of excess O_{2} is very different compared to traditional combustion in gas turbines, which have much larger amount of excess O_{2}. It is assumed that it is possible to have the combustion under near stoichiometric conditions while the emissions of , CO and unburned hydrocarbons are within given constraints. Sundkvist et al. found that using excess of 0.5% of O_{2}, , resulted in 400 ppmv of CO at the turbine outlet [53] and increasing the O_{2} ratio resulted in reduced levels of CO, while increasing the energy penalty from the ASU, as expected.
The pressure drop in the combustion chamber is assumed to be 4%. And a compressor is used to increase the pressure of the fuel above the pressure in the combustion chamber.
2.4. Air Separation Unit
O_{2} is produced with an air separation unit (ASU). The ASU is assumed to be a cryogenic air separation plant. Modelling of the ASU is not within the scope of this current study. ASU power consumption is highly dependent on the purity of the O_{2} stream. It is therefore an economic tradeoff between purity and cost. Typical stateoftheart cryogenic ASU can produce O_{2} with 99.5%volume purity at a power consumption of 900 kW/(kg/s) [54]. By decreasing the purity, it is possible to reduce the power consumption of the ASU. At a purity level of 95%, the power consumption can be assumed to be around 735 kW/(kg/s) [17, 55]. In this study this has been taken into account and a purity level of 95% for the ASU is used. The corresponding O_{2} composition is shown in Table 4.

The ASU unit delivers the O_{2} stream at a pressure of 1.2 bar and with a temperature of 30°C. An intercooled compressor is used to increase the pressure of the stream to the working pressure in the combustor. The compression process has been modelled in the cycle simulation.
2.5. Flue Gas Condenser
The main purpose of the oxyfuel combustion cycles is to produce CO_{2} along with power generation. Because the flue gas consists mostly of CO_{2} and steam, the most convenient method is to condense the water from the flue gas to produce the CO_{2}. The flue gas will also contain small amounts of Ar, N_{2}, and O_{2}. There will also be traces of harmful acid gases along with particles such as soot. By using a direct contact condenser, these harmful gases and the particles can be removed from the flue gas when the steam is condensed. The condenser will therefore also act as a scrubber.
The efficiency of the condenser is defined aswhere is the amount of water that is condensed from the flue gas and is the amount of water in the flue gas that enters the condenser. The flue gas condenser efficiency is a simple way to evaluate the performance of the condensers [56]. The parameter does not represent an efficiency in its true sense but is a metric commonly used to describe the performance of condensers [56].
2.6. CO_{2} Compression
The CO_{2} stream from the condenser that will be sent to storage needs to be compressed to a higher pressure and the remaining water vapour and noncondensable gases need to be removed. This process, the CO_{2} recovery and compression process, is not within the system boundaries of the current study. It is instead taken into account by assuming a fixed energy cost, kW/(kg/s) of wet CO_{2} [25]. This energy cost assumes that the stream is compressed to 100 bar. This value also takes into account the removal of water and other gases that are present in the CO_{2} stream.
3. Power Cycles
The fuel is assumed to be natural gas and the composition is shown in Table 5. Common assumptions used in the simulations of the cycles are shown in Table 6.


3.1. Reference Cycle
A reference cycle was modelled that is in the midsize range. The midsize range is from 30 to 150 MW [42]. Here we have aimed at keeping the gross combined power output from the gas turbine and the steam turbine constant at 100 MW. The reference cycle has been modelled as a gas generator and a separate power turbine, that is, a twoshaft gas turbine. The gas generator turbine consists of two cooled stages. The cooling flow is bled from the compressor. The steam cycle for a power plant in this power range usually employs single or double pressure levels and does not use reheat. Here we have used a dualpressure steam cycle without reheat. The steam turbine is a singlecasing nonreheat. The pressure was set to 140 bar and the maximum temperature to 560°C at the inlet to the steam turbine. If the exhaust temperature from the gas turbine goes below 585°C the steam temperature decreases so that the temperature difference is 25°C. A schematic of the cycle is shown in Figure 1.
3.2. SCOCCC
A schematic of the SCOCCC is shown in Figure 2. The SCOCCC is based on the reference cycle. Now, however, the fuel is combusted with O_{2} that is produced in the ASU. The fuel is combusted near to stoichiometric ratio, meaning that nearly no excess O_{2} is produced. This minimizes the power demand of the ASU. The combustion chamber is cooled using the recycled flue gas, after most of the steam is condensed from it, in the condenser. The flue gas leaving the combustion chamber is mainly CO_{2} and also a small amount of steam. The gas turbine layout is the same as the reference cycle with a gas generator and a power turbine. The turbine in the gas generator has two stages, which are both cooled. The cooling flow is also bled from the compressor, similar to the reference cycle. The layout of the steam cycle is unchanged from the reference cycle. The flue gas goes to the condenser after the heat recovery steam generator, where the major part of the steam is condensed from the exhaust gas. The flue gas is cooled in this process. The CO_{2} stream that leaves the condenser has near 100% relative humidity. A small part of the CO_{2} stream is sent to the compression and purification process and is then transferred to the storage site. The major part of the CO_{2} is recycled back to the compressor. The water in the CO_{2} stream can possibly condense at the entry to the compressor, which could have a deteriorating effect for the compressor. The CO_{2} stream is therefore heated before it enters the compressor using the heat from the flue condensation.
3.3. Graz Cycle
The main features are that the gas turbine cooling is implemented with steam and that the flue gas is sent straight to the compressor after the HRSG without condensing the steam from it. Part of the flue gas is sent to a condenser where a major part of the water is condensed from it; after this it is sent to the CO_{2} compression and purification process. The CO_{2} is afterwards transferred to the storage site.
The most common layout of the Graz cycle incorporates two bottoming cycles. The first one uses a typical HRSG and a steam turbine, which only expands, however, to the pressure of the combustion chamber. This is because the steam is used for cooling both the combustion chamber and the gas turbine blades. The second bottoming cycle uses the enthalpy of the condensation and assumes that the pressure at the outlet of the condenser is 0.021 bar, which is particularly low.
It is hard to imagine that the first design of the Graz cycle will deviate so greatly from the current layout of the combined cycle. Here we have taken the reference cycle as the basis and implemented the major design features of the Graz cycle. A schematic of the Graz cycle is shown in Figure 3. The cycle incorporates an intercooler to reduce the temperature of the gas at the exit of the compressor as well as steam cooling. This layout, not implementing the second bottoming cycle, is considered more reasonable for the first generation design of the cycle. It also makes the complexity level of the SCOCCCC and the Graz cycle more comparable, by not including improvements that could be implemented on both cycles.
The cycle illustrated in Figure 3 should therefore be understood as a simplified variant of the Graz cycle.
4. Results
A parametric study of the two oxyfuel cycles was performed by varying the turbine entry temperature (TET) and the pressure ratio (PR) of the gas turbine. The turbine entry temperature is the temperature at the exit of the combustion chamber and is therefore also the temperature at the entry to the first stator in the gas turbine. The temperature has been varied from 1250°C to 1600°C. The pressure ratio was varied freely until the design constraints were attained.
4.1. Reference Cycle
The results for the cycle net efficiency are shown in Figure 4 as a function of pressure ratio. The turbine entry temperature is also shown in Figure 4. The net efficiency takes into account the power needed for the pumps in the cycle. The entry temperature for the power turbine has been set to 850°C, to eliminate the need for cooling in the power turbine. If the temperature goes above 850°C, which is the metal temperature limit for the blades, then the first stage in the power turbine would then need to be cooled.
Figure 5 shows the cooling mass flow ratio for the reference cycle. The ratio is defined as the total cooling mass flow divided by the inlet mass flow to the turbine.
4.2. SCOCCC
Figure 6 shows the gross efficiency for the SCOCCC as a function of pressure ratio and turbine entry temperature. The gross efficiency is the total power delivered by the gas turbine and steam turbine generators divided by the energy content of the fuel, based on the lower heating value.
As the pressure ratio decreases, the amount of steam in the low pressure steam is also reduced. The lower limit for the pressure ratio is reached when the mass flow of the low pressure steam approaches zero. The higher pressure ratio limit is reached when the temperature difference for the high pressure steam and the flue gas in the preheater approaches 5°C.
Figure 7 shows the net efficiency for the SCOCCC as a function of pressure ratio and turbine entry temperature. The net efficiency takes into account the fuel compressor power, the power needed for the pumps, the energy needed for the production of the O_{2}, the O_{2} compressor power, and the power needed to compress the CO_{2}. The largest decrease in the efficiency comes from the power requirement for the O_{2} production and compression. As the pressure ratio is increased, the O_{2} compression power consumption increases very rapidly. This results in there being an optimum pressure ratio.
Figure 8 shows the power entry temperature as a function of pressure ratio and turbine entry temperature for the SCOCCC. As can be seen in Figure 8 the power turbine entry temperature for all cases is above the blade material temperature limit, 850°C. This means that the first stage in the power turbine needs to be cooled.
Figure 9 shows the cooling mass flow ratio as a function of pressure ratio and turbine entry temperature for the SCOCCC. The cooling mass flow ratio is higher for the SCOCCC than the reference cycle, since the heat capacity for the working fluid is lower in the SCOCCC than in the reference cycle.
4.3. Graz Cycle
The Graz cycle was studied at turbine entry temperatures of 1250°C, 1450°C, and 1600°C. Figure 10 shows the gross efficiency for the Graz cycle as a function of pressure ratio and turbine entry temperature. It can be seen in Figure 10 that there is no global optimum for the gross efficiency.
Figure 11 shows the net efficiency for the Graz cycle as a function of pressure ratio and turbine entry temperature. The net efficiency is calculated as is done for the SCOCCC. The major reduction in the efficiency comes from the power needed for the O_{2} production and compression. The relative power consumption of the O_{2} compression increases as the pressure ratio increases, which results in an optimum in the net efficiency. Figure 12 shows the power entry temperature as a function of pressure ratio and turbine entry temperature for the Graz cycle. When the power turbine entry temperature is over 850°C, the first stage in the power turbine is cooled.
Figure 13 shows the cooling mass flow ratio as a function of pressure ratio and turbine entry temperature for the Graz cycle. Since steam is used as coolant for the turbine blade cooling for the Graz cycle, the cooling mass flow ratio is considerably lower than for the reference cycle and the SCOCCC. This is the result of the fact that the steam has a substantially lower temperature than the compressor discharge temperature and that the steam has a higher heat capacity than the working fluid.
4.4. Optimum Cycles
The results for the cycles with the optimum performance are shown in Table 7. The optimum reference cycle is determined to be a cycle where there is no need to employ cooling for the power turbine. This means that the power turbine entry temperature is 850°C or lower. The pressure ratio that gives the optimal efficiency is 26.15 and the turbine entry temperature is 1400°C, which results in a net efficiency of 56%. The turbine exhaust temperature is only 526°C, which results in the high pressure steam having a temperature of 501°C since the pinch temperature difference has a minimum value of 25°C in the high pressure (HP) superheater.
 
Same as combustor outlet temperature (COT). Turbine inlet temperature based on the ISO definition. 
The optimum SCOCCC has a relatively high pressure ratio, or around 57.3, and the turbine entry temperature is 1450°C. Even though the pressure ratio is so high, the compressor outlet temperature is only 474°C, which is below the temperature limit of the blade material. The steam turbine produces more of the power in the SCOCCC compared to the reference cycle. The exhaust gas is cooled down from 618°C to 65°C in the HRSG. This comes from the fact that the working fluid achieves a better fit to the steam cycle. The main decrease in power comes from the O_{2} production and the O_{2} compression. The gross efficiency for the cycle is close to 60% but, taking into account the O_{2} production and compression, CO_{2} compression, and also the pumps, this is lowered to 46%. The SCOCCC cycle produces 10.3 kg/s of CO_{2}, which is about 890 tonnes per day. The SCOCCC also produces about 170 kg/s of water with a temperature of 46°C in the flue gas condenser.
The optimum Graz cycle has a pressure ratio of about 36.5 and a turbine entry temperature of 1450°C. Even though the pressure ratio is lower in the Graz cycle than in the SCOCCC and the compression is intercooled, the compressor outlet temperature is much higher or around 605°C. The reason for such a high temperature is mainly the fact that the compressor inlet temperature is 100°C. The working fluid saturation temperature is around 95°C so, to avoid condensation at the inlet of the compressor, the temperature needs to be higher than the saturation temperature. To be able to withstand the high temperature at the outlet of the compressor, the blade material will be more expensive than is normally used in compressors. The gas turbine produces a larger share of the power compared to the reference cycle and the SCOCCC. This is because the cooling in the gas turbine uses steam from the steam cycle. The amount of steam needed for cooling is around 15.4 kg/s, which is about 50% of the steam produced in the steam cycle. This steam will therefore be expanded in the gas turbine and not in the steam turbine. This will have a negative effect on the efficiency since the steam will be expanded to 1 bar instead of 0.045 bar as it is in the steam turbine. One aspect of the Graz cycle is that the power density is much higher compared to both the reference cycle and the SCOCCC. The compressor inlet mass flow is only 40% of the reference cycle mass flow and 50% of that of the SCOCCC. The gross efficiency for the Graz cycle is around 59%. The major deduction in efficiency comes from the O_{2} production and compression. However, the compression power consumption is lower in the Graz cycle than the SCOCCC because of the lower pressure ratio. The CO_{2} compression is similar to that in the SCOCCC cycle. This results in nearly the exact same net efficiency as for the SCOCCC or 46%. The Graz cycle produces slightly more CO_{2}, or around 10.8 kg/s, which is about 933 tonnes/day. The Graz cycle produces significantly more water than the SCOCCC, or about 270 kg/s of water with a temperature of 60°C in the flue gas condenser.
5. Discussion and Conclusion
The study compared three combined cycles, a conventional cycle, the SCOCCC, and the Graz cycle, at the midsize level power output. The gross power output for all cycles was set to 100 MW. The conventional cycle was used as the basis for the modelling and as a reference for the oxyfuel combustion cycles. A detailed literature review was conducted for the oxyfuel combustion combined cycles. The literature review showed that there is no consensus on the power requirement for the air separation unit. It also showed that the comparison of the SCOCCC and the Graz cycle has lacked consistent assumptions and agreement on the technology parameters used to model the cycles.
A parametric study was conducted by varying the pressure ratio and the turbine entry temperature for the cycles. A constraint for the conventional cycle was set on the power turbine entry temperature to eliminate the need for cooling in the power turbine. The resulting optimal conventional cycle achieved a 56% net efficiency at a pressure ratio of 26.2 and a turbine entry temperature of 1400°C. The optimal SCOCCC achieved only a 46% net cycle efficiency at a pressure ratio of 57.3 and a turbine entry temperature of 1450°C. The optimal Graz cycle also achieved a net cycle efficiency of 46% at a pressure ratio of 36.5 and a turbine entry temperature of 1450°C. The main reduction in efficiency for the oxyfuel cycles comes from the O_{2} production, which reduced the power output from the cycles by more than 10 MW. An additional reduction of the power output comes from the compression of O_{2} to operating pressure. This is about 7 MW and 6 MW for the SCOCCC and Graz cycle, respectively. The difference comes from the higher pressure ratio of the SCOCCC.
One of the benefits of the Graz cycle is the high power density of the gas turbine. This results in smaller turbomachinery for the gas turbine in the Graz cycle, which lowers the cost of this machinery. One of the main penalties of the Graz cycle is that the large amount of steam, which is generated in the HRSG, is not expanded in the steam turbine but in the gas turbine. The result is that the steam does not expand to the condenser pressure of the steam cycle.
The SCOCCC is considerably simpler than the Graz cycle as it does not implement steam cooling and does not require an intercooler. The optimal SCOCCC, however, has a much higher pressure ratio than both the reference cycle and the Graz cycle. The efficiency does not vary greatly with the pressure ratio, however, and it is possible to reduce the pressure ratio without significantly penalizing the net efficiency. This would facilitate the compressor design substantially.
Nomenclature
:  Hot gas convective heat transfer coefficient 
:  Dimensionless mass flow cooling 
:  Cooling mass flow 
:  Cooling efficiency 
:  Cycle net efficiency 
:  Polytropic efficiency 
:  Condenser efficiency 
:  Ratio of oxygen 
:  Hot gas density 
:  Stanton number of the hot gas 
:  Cooling effectiveness 
:  Coolant mass flow ratio 
:  Blade area 
:  Annulus area 
:  Heat capacity of the cooling fluid 
:  Heat capacity of the hot gas 
:  Pressure 
PR:  Pressure ratio 
:  Gas constant 
:  Turbine loss parameter 
:  Entropy 
:  Uniform blade temperature 
:  Cooling flow exit temperature 
:  Cooling flow inlet temperature 
:  Hot gas temperature 
TET:  Turbine entry temperature 
TIT:  Turbine inlet temperature 
ASU:  Air separation unit 
CCS:  Carbon Capture and Storage 
GT:  Gas turbine 
SCOCCC:  Semiclosed oxyfuel combustion combined cycle 
ST:  Steam turbine. 
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
This research was funded by the Swedish Energy Agency, Siemens Industrial Turbomachinery AB, GKN Aerospace, and the Royal Institute of Technology through the Swedish research program TURBOPOWER. Their support is gratefully acknowledged. The financial grant from Landsvirkjun’s Energy Research Fund is gratefully acknowledged by the first author.
References
 IPCC, “2013: Summary for policymakers,” in Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, T. Stocker, D. Qin, G.K. Plattner et al., Eds., Cambridge University Press, Cambridge, UK, 2013. View at: Google Scholar
 B. Metz, O. Davidson, H. de Coninck, M. Loos, and L. Meyer, “IPCC special report on carbon dioxide capture and storage,” Tech. Rep., Intergovernmental Panel on Climate Change, Working Group III, Geneva, Switzerland, 2005. View at: Google Scholar
 F. Bolland, R. Naqvi, R. Span et al., “Public summary report of ENCAP delivarables D6.1.4 evaluation of technologies and benchmarking based on reference cases and D6.2.1 modelling, design and operational analysis,” Tech. Rep., 2007, http://www.encapco2.org/. View at: Google Scholar
 O. Bolland and S. Sæther, “New concepts for natural gas fired power plants which simplify the recovery of carbon dioxide,” Energy Conversion and Management, vol. 33, no. 5–8, pp. 467–475, 1992. View at: Publisher Site  Google Scholar
 I. Ulizar and P. Pilidis, “A semiclosedcycle gas turbine with carbon dioxideargon as working fluid,” Journal of Engineering for Gas Turbines and Power, vol. 119, no. 3, pp. 612–616, 1997. View at: Publisher Site  Google Scholar
 I. Ulizar and P. Pilidis, “Design of a semiclosedcycle gas turbine with carbon dioxide: argon as working fluid,” Journal of Engineering for Gas Turbines and Power, vol. 120, no. 2, pp. 330–335, 1998. View at: Publisher Site  Google Scholar
 I. Ulizar and P. Pilidis, “Handling of a semiclosed cycle gas turbine with a carbon dioxideargon working fluid,” Journal of Engineering for Gas Turbines and Power, vol. 122, no. 3, pp. 437–441, 2000. View at: Publisher Site  Google Scholar
 O. Bolland and P. Mathieu, “Comparison of two CO_{2} removal options in combined cycle power plants,” Energy Conversion and Management, vol. 39, no. 16–18, pp. 1653–1663, 1998. View at: Publisher Site  Google Scholar
 J.M. Amann, M. Kanniche, and C. Bouallou, “Natural gas combined cycle power plant modified into an O_{2}/CO_{2} cycle for CO_{2} capture,” Energy Conversion and Management, vol. 50, no. 3, pp. 510–521, 2009. View at: Publisher Site  Google Scholar
 S. H. Tak, S. K. Park, T. S. Kim, J. L. Sohn, and Y. D. Lee, “Performance analyses of oxyfuel power generation systems including CO_{2} capture: comparison of two cycles using different recirculation fluids,” Journal of Mechanical Science and Technology, vol. 24, no. 9, pp. 1947–1954, 2010. View at: Publisher Site  Google Scholar
 K. Jordal, O. Bollard, and A. Klang, “Aspects of cooled gas turbine modeling for the semiclosed O_{2}/CO_{2} cycle with CO_{2} capture,” Journal of Engineering for Gas Turbines and Power, vol. 126, no. 3, pp. 507–515, 2004. View at: Publisher Site  Google Scholar
 R. E. Ulfsnes, O. Bolland, and K. Jordal, “Modelling and simulation of transient performance of the semiclosed O_{2}/CO_{2} gas turbine cycle for CO_{2}capture,” in Proceedings of the ASME Turbo Expo: Power for Land, Sea and Air, GT200338068, Atlanta, Ga, USA, June 2003. View at: Publisher Site  Google Scholar
 R. Ulfsnes, G. Karlsen, K. Jordal, O. Bolland, and H. M. Kvamsdal, “Investigation of physical properties of CO_{2}/H_{2}Omixtures for use in semiclosed O_{2}/CO_{2} gas turbine cycle with CO_{2}capture,” in Proceedings of the 16th International Conference on Efficiency, Cost, Optimization, Simulation, and Environmental Impact of Energy Systems (ECOS '03), Copenhagen, Denmark, June 2003. View at: Google Scholar
 G. Corchero, V. P. Timón, and J. L. Montañés, “A natural gas oxyfuel semiclosed combined cycle for zero CO_{2} emissions: a thermodynamic optimization,” Proceedings of the Institution of Mechanical Engineers Part A: Journal of Power and Energy, vol. 225, no. 4, pp. 377–388, 2011. View at: Publisher Site  Google Scholar
 T. Riethmann, F. Sander, and R. Span, “Modelling of a supercharged semiclosed oxyfuel combined cycle with CO_{2} capture and analysis of the partload behavior,” Energy Procedia, vol. 1, no. 1, pp. 415–422, 2009. View at: Publisher Site  Google Scholar
 H. J. Yang, D. W. Kang, J. H. Ahn, and T. S. Kim, “Evaluation of design performance of the semiclosed oxyfuel combustion combined cycle,” Journal of Engineering for Gas Turbines and Power, vol. 134, Article ID 111702, 2012. View at: Publisher Site  Google Scholar
 A. Dahlquist, M. Genrup, M. Sjoedin, and K. Jonshagen, “Optimization of an oxyfuel combined cycle regarding performance and complexity level,” in Proceedings of the ASME Turbo Expo 2013: Power for Land, Sea and Air, GT201394755, American Society of Mechanical Engineers, San Antonio, Tex, USA, 2013. View at: Google Scholar
 M. Sammak, E. Thorbergsson, T. Grönstedt, and M. Genrup, “Conceptual meanline design of single and twinshaft oxyfuel gas turbine in a semiclosed oxyfuel combustion combined cycle,” Journal of Engineering for Gas Turbines and Power, vol. 135, no. 8, Article ID 081502, 2013. View at: Publisher Site  Google Scholar
 H. Jericha, “Efficient steam cycles with internal combustion of hydrogen and stoichiometric oxygen for turbines and piston engines,” International Journal of Hydrogen Energy, vol. 12, no. 5, pp. 345–354, 1987, CIMAC Conference Paper, Oslo, Norway. View at: Publisher Site  Google Scholar
 H. Jericha, W. Sanz, J. Woisetschläger, and M. Fesharaki, “CO_{2}retention capability of CH_{4}/O_{2}fired graz cycle,” in Proceedings of the CIMAC World Congress on Combustion Engine Technology, Paper G07, pp. 1–13, Interlaken, Switzerland, 1995. View at: Google Scholar
 H. Jericha and E. Göttlich, “Conceptual design for an industrial prototype Graz cycle power plant,” in Proceedings of the ASME Turbo Expo 2002: Power for Land, Sea and Air, GT200230118, American Society of Mechanical Engineers, Amsterdam, The Netherlands, June 2002. View at: Google Scholar
 F. Heitmeir, W. Sanz, E. Göttlich, and H. Jericha, “The graz cyclea zero emission power plant of highest efficiency,” in XXXV Kraftwerkstechnisches Kolloquium, Dresden, Germany, September 2003. View at: Google Scholar
 H. Jericha, E. Göttlich, W. Sanz, and F. Heitmeir, “Design optimization of the graz cycle prototype plant,” Journal of Engineering for Gas Turbines and Power, vol. 126, no. 4, pp. 733–740, 2004. View at: Publisher Site  Google Scholar
 F. Heitmeir and H. Jericha, “Turbomachinery design for the Graz cycle: an optimized power plant concept for CO_{2} retention,” Proceedings of the Institution of Mechanical Engineers Part A: Journal of Power and Energy, vol. 219, no. 2, pp. 147–155, 2005. View at: Publisher Site  Google Scholar
 W. Sanz, H. Jericha, F. Luckel, E. Göttlich, and F. Heitmeir, “A further step towards a Graz cycle power plant for CO_{2} capture,” in Proceedings of the ASME Turbo Expo: Power for Land, Sea and Air, GT200568456, pp. 181–190, American Society of Mechanical Engineers, Reno, Nev, USA, June 2005. View at: Publisher Site  Google Scholar
 W. Sanz, H. Jericha, M. Moser, and F. Heitmeir, “Thermodynamic and economic investigation of an improved Graz Cycle power plant for CO_{2} capture,” Journal of Engineering for Gas Turbines and Power, vol. 127, no. 4, pp. 765–772, 2005. View at: Publisher Site  Google Scholar
 M. H. Jericha, M. W. Sanz, and M. E. Goettlich, “Gasturbine with CO_{2} retention—400 MW oxyfuelsystem Graz cycle,” in Proceedings of the CIMAC World Congress on Combustion Engine Technology, Vienna, Austria, May 2007. View at: Google Scholar
 H. Jericha, W. Sanz, and E. Göttlich, “Design concept for large output graz cycle gas turbines,” Journal of Engineering for Gas Turbines and Power, vol. 130, no. 1, Article ID 011701, 2008. View at: Publisher Site  Google Scholar
 H. Jericha, W. Sanz, E. Göttlich, and F. Neumayer, “Design details of a 600 mw Graz cycle thermal power plant for CO_{2} capture,” in Proceedings of the ASME Turbo Expo: Power for Land, Sea and Air, GT200850515, pp. 507–516, American Society of Mechanical Engineers, Berlin, Germany, June 2008. View at: Publisher Site  Google Scholar
 H. Kvamsdal, O. Maurstad, K. Jordal, and O. Bolland, “Benchmarking of gasturbine cycles with CO_{2} capture,” in Proceedings of the 7th International Conference on Greenhouse Gas Control Technologies, vol. 1, pp. 233–242, Vancouver, Canada, September 2004. View at: Google Scholar
 F. Franco, T. Mina, G. Woolatt, M. Rost, and O. Bolland, “Characteristics of cycle components for CO_{2} capture,” in Proceedings of the 8th International Conference on Greenhouse Gas Control Technologies, Trondheim, Norway, June 2006. View at: Google Scholar
 H. M. Kvamsdal, K. Jordal, and O. Bolland, “A quantitative comparison of gas turbine cycles with CO_{2} capture,” Energy, vol. 32, no. 1, pp. 10–24, 2007. View at: Publisher Site  Google Scholar
 W. Sanz, H. Jericha, B. Bauer, and E. Göttlich, “Qualitative and quantitative comparison of two promising oxyfuel power cycles for CO_{2} capture,” Journal of Engineering for Gas Turbines and Power, vol. 130, no. 3, Article ID 031702, 2008. View at: Publisher Site  Google Scholar
 G. Woollatt and F. Franco, “Natural gas oxyfuel cyclespart 1: conceptual aerodynamic design of turbomachinery components,” Energy Procedia, vol. 1, pp. 573–580, 2009. View at: Google Scholar
 E. Thorbergsson, T. Grönstedt, M. Sammak, and M. Genrup, “A comparative analysis of two competing midsize oxyfuel combustion cycles,” in Proceedings of the ASME Turbo Expo: Power for Land, Sea and Air, GT201269676, pp. 375–383, American Society of Mechanical Engineers, Copenhagen, Denmark, June 2012. View at: Publisher Site  Google Scholar
 P. Mathieu and O. Bolland, “Comparison of costs for natural gas power generation with CO_{2} capture,” Energy Procedia, vol. 37, pp. 2406–2419, 2013. View at: Google Scholar
 O. Bolland and H. Undrum, “A novel methodology for comparing CO_{2} capture options for natural gasfired combined cycle plants,” Advances in Environmental Research, vol. 7, no. 4, pp. 901–911, 2003. View at: Publisher Site  Google Scholar
 SimTech Simulation Technology, User Documentation: Program Modules and Model Libraries, IPSEpro Process Simulator, 2003.
 E. Thorbergsson, T. Grönstedt, and C. Robinson, “Integration of fluid thermodynamic and transport properties in conceptual turbomachinery design,” in Proceedings of the ASME Turbo Expo: Power for Land, Sea and Air, GT201395833, American Society of Mechanical Engineers, San Antonio, Tex, USA, June 2013. View at: Publisher Site  Google Scholar
 G. Halls, “Air cooling of turbine blades and vanes: an account of the history and development of gas turbine cooling,” Aircraft Engineering and Aerospace Technology, vol. 39, no. 8, pp. 4–14, 1967. View at: Publisher Site  Google Scholar
 M. J. Holland and T. F. Thake, “Rotor blade cooling in high pressure turbines,” Journal of Aircraft, vol. 17, no. 6, pp. 412–418, 1980. View at: Publisher Site  Google Scholar
 K. Jordal, Modeling and performance of gas turbine cycles with various means of blade cooling [Ph.D. thesis], Department of Heat and Power Engineering, Lund University, 2001.
 O. Bolland and J. F. Stadaas, “Comparative evaluation of combined cycles and gas turbine systems with water injection, steam injection, and recuperation,” Journal of Engineering for Gas Turbines and Power, vol. 117, no. 1, pp. 138–145, 1995. View at: Publisher Site  Google Scholar
 M. A. ElMasri, “On thermodynamics of gasturbine cycles: Part 2—a model for expansion in cooled turbines,” Journal of Engineering for Gas Turbines and Power, vol. 108, no. 1, pp. 151–159, 1986. View at: Publisher Site  Google Scholar
 K. Jordal, L. Torbidoni, and A. Massardo, “Convective blade cooling modelling for the analysis of innovative gas turbine cycles,” in Proceedings of the ASME Turbo Expo 2001: Power for Land, Sea and Air, GT20010390, American Society of Mechanical Engineers, New Orleans, La, USA, 2001. View at: Google Scholar
 J. H. Horlock, D. T. Watson, and T. V. Jones, “Limitations of gas turbine performance imposed by large turbine cooling flows,” Journal of Engineering for Gas Turbines and Power, vol. 123, no. 3, pp. 487–494, 2001. View at: Publisher Site  Google Scholar
 K. Jordal, Gas turbine cooling modeling—thermodynamic analysis and cycle simulations [Licentiate thesis], Department of Heat and Power Engineering, Lund University, Lund, Sweden, 1999.
 R. C. Wilcock, J. B. Young, and J. H. Horlock, “The effect of turbine blade cooling on the cycle efficiency of gas turbine power cycles,” Journal of Engineering for Gas Turbines and Power, vol. 127, no. 1, pp. 109–120, 2005. View at: Publisher Site  Google Scholar
 M. Mallen and G. Saville, Polytropic Processes in the Performance Prediction of Centrifugal Compressors, I. Mech. E. Conference Publications, Institution of Mechanical Engineers, London, UK, 1977.
 W. Traupel, Thermische Turbomaschinen, Springer, Berlin, Germany, 1977.
 F. Di Maria and K. Jordal, “Blade cooling model comparison and thermodynamic analysis,” in Proceedings of the 5th ASME/JSME Joint Thermal Engineering Conference, AJTE996117, American Society of Mechanical Engineers, San Diego, Calif, USA, March 1999. View at: Google Scholar
 A. Dahlquist, M. Thern, and M. Genrup, “The inuence from the working medium on the profile loss in compressor and turbine airfoils,” in Proceedings of the ASME Turbo Expo 2014: Power for Land, Sea and Air, GT201425069, American Society of Mechanical Engineers, Düsseldorf, Germany, 2014. View at: Google Scholar
 S. G. Sundkvist, A. Dahlquist, J. Janczewski et al., “Concept for a combustion system in oxyfuel gas turbine combined cycles,” Journal of Engineering for Gas Turbines and Power, vol. 136, no. 10, Article ID 101513, 2014. View at: Publisher Site  Google Scholar
 I. Pfaff and A. Kather, “Comparative thermodynamic analysis and integration issues of CCS steam power plants based on oxycombustion with cryogenic or membrane based air separation,” Energy Procedia, vol. 1, pp. 495–502, 2009. View at: Google Scholar
 J. Tranier, N. Perrin, and A. Darde, “Update on advanced developments for ASU and CO_{2} purification units for oxycombustion (air liquide, France),” in Proceedings of the 3rd Meeting of the OxyFuel Combustion Network, IEAGHG International OxyCombustion Network, Yokohama, Japan, March 2008. View at: Google Scholar
 K. Jeong, M. J. Kessen, H. Bilirgen, and E. K. Levy, “Analytical modeling of water condensation in condensing heat exchanger,” International Journal of Heat and Mass Transfer, vol. 53, no. 1112, pp. 2361–2368, 2010. View at: Publisher Site  Google Scholar
Copyright
Copyright © 2016 Egill Thorbergsson and Tomas Grönstedt. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.