Talanta 132 (2015) 327–333
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Effective identification of (NH4)2CO3 and NH4HCO3 concentrations in NaHCO3 regeneration process from desulfurized waste Muthuraman Govindan 1, Kannan Karunakaran 1, Palanisami Nallasamy 1, Il Shik Moon n Department of Chemical Engineering, Sunchon National University, 315 Maegok Dong, Suncheon 540-742, Chonnam, South Korea
art ic l e i nf o
a b s t r a c t
Article history: Received 6 June 2014 Received in revised form 13 August 2014 Accepted 14 August 2014 Available online 28 August 2014
This work describes the quantitative analysis of (NH4)2CO3 and NH4HCO3 using a simple solution phase titration method. Back titration results at various (NH4)2CO3–NH4HCO3 ratios demonstrated that 6:4 ratio caused a 3% error in their differentiation, but very high errors were found at other ratios. A similar trend was observed for the double indicator method, especially when strong acid HCl was used as a titrant, where still less errors (2.5%) at a middle ratio of (NH4)2CO3–NH4HCO3 was found. Remaining ratios with low (NH4)2CO3 (2:8, 4:6) show high þve error (found concentration is less) and high (NH4)2CO3 (7:3, 8:2, and 9:1) show high –ve error (found concentration is higher) and vice versa for NH4HCO3. In replacement titration using Na2SO4, at both higher end ratios of (NH4)2CO3–NH4HCO3 (2:8 and 9:1), both –ve and þve errors were minimized to 75% by partial equilibrium arrest between (NH4)2CO3 and NH2COONH4, instead of more than 100% observed in back titration and only double indicator methods. In the presence of (NH4)2SO4 both –ve and þve error% are completely reduced to 3 7 1 at ratios 2:8, 4:6, and 6:4 of (NH4)2CO3–NH4HCO3, which demonstrates that the equilibrium transformation between NH2COONH4 and (NH4)2CO3 is completely controlled. The titration conducted at lower temperature (5 1C) in the presence of (NH4)2SO4 at higher ratios of (NH4)2CO3–NH4HCO3 (7:3, 8:2, and 9:1) shows complete minimization of both –ve and þve errors to 2 71%, which explains the complete arresting of equilibrium transformation. Finally, the developed method shows 27 1% error in differentiation of CO23 and HCO3 in the regeneration process of NaHCO3 from crude desulfurized sample. The developed method is more promising to differentiate CO23 and HCO3 in industrial applications. & 2014 Elsevier B.V. All rights reserved.
Keywords: Flue Gas Desulfurization Sodium sulfate Ammonium sulfate Ammonium carbonate Ammonium bicarbonate Titration
1. Introduction Growing awareness of the risks of climate change has generated public concerns, and over the last two decades the sequestering of air pollutants has been of major research interest. As the world's population and energy demand increase, continued industrialization will undoubtedly increase levels of atmospheric pollutants like NOx, SOx, and CO2. In particular power plant and steel plant furnaces emit SOx and NOx (called flue gases), which are removed using wet or dry sorbent methods [1]. Flue Gas Desulfurization (FGD) is performed using the wet sorbent method. However, due to the large amount of space required for installation, the large volume of water required and high capital and operating costs [2], research has shifted toward dry methods of SO2 removal. Various types of solid sorbent/catalysts are being used for dry FGD, such as, calcium based [3], sodium based [4], activated carbon [5],
n
Corresponding author. Tel.: þ 82 61 750 3581; fax: þ 82 61 750 3581. E-mail address: [email protected] (I.S. Moon). 1 These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.talanta.2014.08.044 0039-9140/& 2014 Elsevier B.V. All rights reserved.
metal oxide [6], and zeolite sorbents [7], waste-derived siliceous materials [8], and mediated electrocatalytic oxidation (MEO) [9,10]. Many of these methods have limitations in terms of their commercialization. Among these methods, sodium-based sorbents are considered more effective for FGD, especially for SOx. These processes are conducted mainly using sodium carbonate (Na2CO3), sodium bicarbonate (NaHCO3), and trona (Na2CO3 NaHCO3 2H2O). In this process, either sodium bicarbonate (NaHCO3) (or) sodium carbonate (Na2CO3) reacts with SO2 to form Na2SO3 which is further converted to sodium sulfate (Na2SO4) in the presence of excess oxygen, as explained by the following equation sequences [1–4]: 2NaHCO3 þSO2-Na2SO3 þ 2CO2 þH2O
(1)
2NaHCO3-Na2CO3 þCO2 þH2O
(2)
Na2CO3 þSO2-Na2SO3 þ CO2
(3)
Na2SO3 þ1/2O2-Na2SO4
(4)
However, it had a few drawbacks: (i) sodium bicarbonate is expensive; (ii) the resulting by-product from the sodium
328
M. Govindan et al. / Talanta 132 (2015) 327–333
NH3storage
Waste solution
CO2storage
Gas
Homogenizer with controlled temperature
Crude desulfurized waste mixture
Separator
Ammonium sulfate secondary waste
Reactor with controlled temperature
L S
L S Applications
Impurities
NaHCO3
Dryer
Fig. 1. Schematic representation for the regeneration pathway of NaHCO3 from desulfurized waste sample.
bicarbonate–SOx reaction (sodium sulfate) is of limited economic value; and (iii) sodium sulfate disposal is expensive and poses significant environmental challenges. To regenerate sodium bicarbonate from secondary waste Na2SO4 (SS) provides an efficient way of avoiding the above mentioned drawbacks [11]. The regeneration of sodium bicarbonate from sodium sulfate is schematically presented in Fig. 1. The regeneration of sodium bicarbonate involves three steps: (i) separation of the sodium sources from their impurities, (ii) regeneration of NaHCO3 (SBC) using NH4HCO3 (ABC) or gases of NH3 and CO2, and (iii) utilization of secondary waste solution (NH4)2SO4 (AS). In these regeneration processes, analysis of each ion plays a crucial role in finalizing the experimental conditions such as the molar ratio of reactants and water content. Particularly the secondary waste AS containg solution, which includes additionally (NH4)2CO3 (AC) and ABC, should be exactly analyzed to perfect the minimization of error. Identifications of CO23 and HCO3 ions are very difficult through analysis by titration due to their complex equilibrium process as follows [5–9,12,13]: þ þ NH4(aq) þ H2O(l)⇌NH3(aq) þH3O(aq)
(5)
þH2O CO23(aq)
(6)
(aq)⇌HCO3(aq) þOH(aq)
þNH4þ (aq)⇌HCO3(aq) þNH3(aq) CO23(aq)
(7)
þNH3(aq)⇌H2NCOO(aq) þH2O(l) HCO3(aq)
(8)
þ þ H2O(l)⇌CO23(aq) þNH4(aq) H2NCOO(aq)
CO23
(9) HCO3
and by a back titration First, the differentiation of method using BaCl2 as a precipitant for CaCO3 has been identified to be infeasible [14]. They found more restriction in minimizing the error in quantification of each compound and no detailed studies about various concentration ratios of CO23 and HCO3 [15]. Later, double indicator Warder's titration method was used to differentiate the CO23 and HCO3 as a next level [16,17]. However, quantifying the CO23 and HCO3 is difficult due to lack of endpoint identification; color change is very feeble in case of AC and ABC [18,19], and no detailed study has been performed by the titration method. In the late 1990s, it was reported that FT-IR, NIR, Raman spectroscopies and powder-XRD can differentiate AC and ABC quantitatively and qualitatively [20–22]. The XRD 2θ peaks of ABC and AC are at 16.5 and 34.4, respectively, and can be used to analyze the solid state compositions of mixtures [21]. In the case of
FT-IR analysis, there is a possibility of quantitative differentiation of ABC (1200 1450 cm 1 for asymmetric CO stretch and 1600 cm 1 for symmetric CO stretch) and AC ( 1300 1600 cm 1 for asymmetric CO stretch). The FT-IR spectra of both solid and solution phases of AC and ABC are quite similar and difficult to distinguish because of broad peaks due to both carbonate and bicarbonate between 1200 and 1700 cm 1 and the same peak at 3098 cm 1, which is due to the ammonium ion [22]. As reported by Meng et al., by variation of oxygen hydrogen (OH) bond of ABC at 1450 nm, which is not present in the carbonate, one can differentiate quantitatively in solid state samples [21]. However, it is difficult to distinguish the bicarbonate from carbonate ion in aqueous solution due to interference of hydroxyl (OH) groups in the water, which restricts the NIR analysis in solution samples. 13C NMR studies have demonstrated the qualitative differentiation of ABC and AC at room temperature [23]. Even the fast equilibrium of HCO3 /CO23 anions results in their two NMR peaks, which permits identifying the HCO3 /CO23 ions quantitatively. Through the Raman spectra, the ABC and AC in solution phase have been differentiated quantitatively using scattered light of the peaks at 1015 cm 1 (for bicarbonate) and 1065 cm 1 (for carbonate) [24]. However, Raman analysis needs an internal standard for exact differentiation of both ABC and AC. The mixture of our secondary waste contains internal standard compound Na2SO4, which should be analyzed additionally using another method. Accordingly several issues must be resolved for the quantitative analyses of ABC and AC in the solution phase, especially in samples obtained during the regeneration of SBC from desulfurized crude waste. There are some models to control the reaction and its coordinates to get a high yield and purity in regeneration of SBC from desulfurized waste [25,26]. Currently there are no reports on controlling the reaction based on analysis of ions present in the crude waste (before reaction) and secondary product waste (after reaction). In the present work a method is developed based on analysis of ions present in crude waste and secondary waste, to control the reaction in the SBC regeneration process. Herein, validations of back titration at different ratios of AC ad ABC were performed for its confirmation. Then, the double indicator method was repeated with different ratios of AC and ABC to check its applicability for their differentiation. Additionally replacement titration, by SBC, was studied. In order to check the equilibrium of CO23 and HCO3 , AS and temperature effect were studied.
M. Govindan et al. / Talanta 132 (2015) 327–333
Finally, the developed method was tested on a real sample from desulfurized crude waste regeneration process of the SBC.
2. Experimental section 2.1. Chemicals Ammonium bicarbonate and ammonium carbonates were purchased from Aldrich. Phenolphthalein, methyl orange, bromocreasol green, phenol red, sodium carbonate, barium chloride dehydrate, sodium hydroxide, and phosphoric acid were purchased from Daejung Chemicals, Korea. Hydrochloric acid and acetic acid were purchased from OCI, and Duksan Korea respectively. 2.2. Reagents 0.1 M stock solutions of hydrochloric acid, acetic acid and phosphoric acid were prepared by dissolving approximately 8.4 ml of 11.6 M hydrochloric acid in 1 L of deionized water, 5.8 ml of 17.4 M acetic acid in 1 L of deionized water, and 6.8 mL of 14.6 M phosphoric acid in 1 L of deionized water respectively. These solutions were standardized by titration against standard sodium carbonate. 2.3. Back titration This method is the oldest method to differentiate two components by precipitating one component [14]. In order to differentiate AC and ABC, various ratios such as (2:8), (4:6), (5:5), (6:4), (7:3), (8:2), (9:1) were chosen. The back titration was performed using two steps. In the first step, total alkalinity was determined by placing a 5 ml sample in a 100 ml conical flask, adding methyl orange indicator and titrating against standardized 0.1 M HCl, as shown in Eq. (10). The end point change from yellow to red was considered as a measure of total alkalinity. This procedure was repeated till concordant values were obtained and the average value was taken. In the second step, 5 ml of sample was taken in a 100 ml conical flask, then 16 ml of NaOH and 0.88 ml of 10% BaCl2 solution were added along with a few drops of phenolphthalein indicator. Instantly, the white precipitate (BaCO3) generated in the sample mixture solution was then immediately titrated against standardized 0.1 M HCl solution after using a high speed centrifuge for a short time (30 s), as shown in Eq. 11. The disappearance of pink color is an indication of end point. The volume of consumed HCl was recorded and the test was repeated for concordant values. Step 1: Total alkalinity: NH2COONH4 and (NH4)2CO3 þ NH4HCO3 þ3HCl-3NH4 Cl þ2H2Oþ2CO2
(10)
329
The same method has been adopted to identify ammonium carbonate and ammonium bicarbonate [27]. Briefly, ammonium carbonate and ammonium bicarbonate mixture was dissolved in 100 ml of water. 5 ml of this solution was transferred into a conical flask. 2 drops of phenolphthalein were added and pink color appeared. When titrated against 0.1 M acids the pink color disappeared, which is the endpoint of ammonium carbonate solution. To the same solution 2 drops of phenol red were added and the solution became yellow. After titration again with acetic acid the yellow turned into red, which is the ammonium bicarbonate endpoint. For phosphoric acid and hydrochloric acid, methyl orange is used for ammonium bicarbonate and the end point is orange to red. In all titration experiments, the titration time was 0.5–3 min for all titrations at all ratios of AC and ABC. In the AC determination step, the time spent is between 0.5 and 1.5 min. The error% of ABC was calculated by substracting the AC value obtained from the phenolpthalein indicator by total alkalinity (ACl) value obtained from the second indicator (methyl orange). As shown in Eqs. (12)–(14). to cross check, the AC value added with ABC, which is equal to carbamic acid (CAm) or carbonic acid (CA) that developed during titration. Then, the CAm or CA was subtracted by ACl to get right ABC. CACl – CAC or CACm ¼ CABC
(12)
CAC þ CABC ¼CCAm
(13)
CACl – CCAm
or CA
or CA ¼CABC
(14)
In the replacement titration, two times the concentration of Na2SO4 was added and repeated as a double indicator method of titration. In stabilization experiments, (NH4)2SO4 and temperature were included as two times higher molar concentration and 25 to 5 1C variation respectively. The room temperature experiments were carried out at 207 2 1C. All the experiments were repeated until the results were reproducible. 2.5. Analysis of real regeneration process samples First, an industrial desulfurized sample 10 g was dissolved in 250 ml of deionized water and filtered to remove impurities, and separated the salt solution for analysis. The Na þ and NH4þ ions were estimated using ion selective electrodes (ISE) that purchased from VAN LONDON Phoenix.CO (No.: NA715XX for Naþ and No.: NH415XX) with the help of iSTEK multimeter (Model No.: pH-240 L) from USA. Prior to the analysis, the electrodes and instruments were calibrated as per the instruction that given by a manufacturing company. In another way, the NH4þ was cross checked with standard procedure by UV-Visible spectroscopy (Shimadzu, UV160A) from Japan, by the indophenol blue method at 630 nm [28]. The sulphate ion in the sample was estimated in the form of Barium sulphate by the addition of Barium chloride precipitating agent along with Conditional reagent followed by UV–vis spectrophotometric absorption at 420 nm [29]. All metal ions that present in the real sample was analyzed by ICP-MS (model ELANDRCe, Perkin Elmer) from USA.
Step 2: Differentiate HCO3 and CO23 2HCl
ðNH4Þ2 CO3 þ NH4 HCO3 þ NaOH þBaCl2 ⟹3NH4 Cl þ BaCO3 þ H2 O þ NaCl þ H2 CO3 ð11Þ
2.4. Double indicator method There are some reports for differentiation of sodium carbonate and bicarbonate in the reaction mixture [16,17], and a few reports for ABC and AC without detailed experimental conditions [15].
3. Result and discussion 3.1. Conventional back titration on differentiation of (NH4)2CO3/ (NH4)HCO3 As evidenced in previous reports, SBC and SC can be differentiated effectively by back titration with an error of less than 5%. However, they never considered ACm equilibrium in back titration to differentiate ABC and AC in CO2 absorption process using ammonia [14]. In the present case, desulfurized waste contains many salts that including SBC and SC in different ratios, which
330
M. Govindan et al. / Talanta 132 (2015) 327–333
3.2. Double indicator method (NH4)2CO3 NH4HCO3
Found difference (g)
0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8
2:8
4:6
5:5
6:4
7:3
8:2
9:1
Concn. by ratio Fig. 2. Effect of different concentration ratios on found concentration differences of (NH4)2CO3 and (NH4)HCO3 in their identification through back titration.
must be accurately identified for further reaction with ABC to regenerate SBC. After reaction or to control the reaction, differentiation of ABC and AC is important. By keeping all in mind, we have started optimizing conditions in back titration such as the effect of NaOH and BaCl2 addition time in particular of AC to ABC ratio 6:4. The obtained results are presented in supporting information Table S1. There appears minimized error% during the both NaOH and BaCl2 addition time in 30 s. Using these selected conditions, the back titrations to identify the ABC and AC in different ratios at room temperature was performed and the obtained results are depicted in Fig.2. At high concentration of AC, the found concentration of AC increased to 0.35 g than (þve error) the actual input value. At the same time, the found ABC is very decreased to 0.54 g (-ve error) than the originally added in 8:2 ratio. The found difference of both ABC and AC increased and decreased in a similar manner with increasing ratio up to 6:4, where the found difference is almost 0.01 g which equal to less than 3% þve error. Further increasing the ratios beyond 6:4 of AC and ABC, the found difference shows a reverse trend which means, the found AC is decreased than the originally added and ABC is increased than the input concentrations. The overall found difference explains that the highest concentration of AC or ABC are tended to convert to another form chemically. The chemical equilibrium may dominate more at high end concentration ratios. It is reported that the ABC is converted to AC at high temperature and excess of CO2 [27]. In a similar way, AC converted to ABC in excess CO2 and H2O [30]. But only containing of both ABC and AC, how the conversion has occurred? Instead of conversion, we suggest the possibility of equilibrium stabilization, as shown in the following eqs. (15 and 16). Another possible way for high error % is the formation of ACm from AC in water as shown in eqs. 8 and 9. It is evidenced that the ACm will not precipitate in precipitation step upon addition of BaCl2 [31] to remove the AC from the titer solution in related to avoid its interference during the back titration of the consumed NaOH with HCl. In case ACm remains in titer solution, the consumed HCl volume will vary which in turn leads high error% in determination of AC, as shown in eq. (17) At High (NH4)2CO3: (NH4)2CO3 "NH4HCO3
(15)
At high NH4HCO3: (NH4)2CO3" NH4HCO3
(16)
(NH4)2CO3 þ NH4HCO3 þHCl-2NH4HCO3 þNH4Cl
(18)
NH2COONH4 þNH4HCO3 þHCl-NH4HCO3 þNH2COOHþNH4Cl (180 ) 2NH4HCO3 þ2HCl-2NH4Cl þ2H2CO3
(19)
NH4 HCO3 þ NH2 COOH þ HCl!NH2 COOH þ H2 CO3 þ NH4 Cl
ð190 Þ
NH2COOHþHCl-NH4ClþCO2
(200 )
2
Depending upon the acid, which we used, the error% is varying for both the ABC and AC or ACm that can be defined as the -ve error% (found value is higher than originally taken) and þ ve error% 0.4 0.0 -0.4 -0.8
(NH4)2CO3 NH4HCO3
0.4 0.0 -0.4 -0.8 0.4 0.0 -0.4 -0.8 2:8
4:6
5:5
6:4
7:3
8:2
9:1
Concn, by ratio
HCL
NH2 COONH4 þ NH4 HCO3 þ NaOH þ BaCl2 !2NH4 Cl þ BaCO3 þ H2 O þ NaCl þ NH2 COOH
Because of defined differences in the pH value of ABC (8.3), and AC (9), the two indicator method could possibly provide accurate ABC and AC concentrations. In consideration of ACm, the pH of ACm 8.9 [32], which is very close to the pH of AC, it clearly helps to differentiate ABC even if it exists because of phenolpthalein used as indicator. Since ABC, ACm, and AC are weak bases, acid selection could minimize errors. We selected three different acids, that is, acetic acid (weak base vs weak acid) and phosphoric and hydrochloric acids (weak base vs strong acid). As evident in literature, In order to know the time dependent equilibrium of AC, ACm, and ABC, the sample titration time at minimum (1 min) and maximum (720 min) were done and the results are tabulated and presented in supporting information Table S2. After 1 min, the error% has increased to only 7 to 10 at AC to ABC of 75:25 ratio. In consideration of practical difficulties, 1 min sampling time was selected for all the titrations. During analysis, we found difficulties in end point measurements, especially for AC and ACm determinations, when only a pale pink was obtained after adding phenolphthalein. This color change behavior just contrary to the SBC and SC differentiation, there was able to judge a clear difference in end point and correct value. The titration results obtained for ABC and AC are depicted in Fig. 3. In plot (b) of Fig.3 illustrates the titration of acetic acid with (NH4)2CO3/(NH4)HCO3 mixture. During titration, AC or ACm becomes ABC or CAm and carbonic acid (CA) at second endpoint while the use of methyl orange indicator as shown in eqs. 18–20’ in case of ACm. The % errors were calculated using eqs. (12–14).
Found difference (g)
0.6
ð17Þ
Fig. 3. Effect of acids on titration of different concentration ratios of (NH4)2CO3/ (NH4)HCO3 mixture in their identification by double indicator method: (a) Phosphoric acid; (b) Acetic acid; (c) Hydrochloric acid.
M. Govindan et al. / Talanta 132 (2015) 327–333
3.3. Identification through replacement titration by Na2SO4 It is known that alkali salts reacts with ABC and AC form alkali carbonate and bicarbonate [11]. Using this approach, sodium sulfate was added accordingly to the concentrations of ABC and AC to convert them to SBC and SC, as shown in the eqs. 21, and 22, this facilitates the accurate differentiation of ABC and AC. Na2SO4 þ(NH4)2CO3-Na2CO3 þ(NH4)2SO4
(21)
Na2SO4 þ2(NH4)HCO3-2NaHCO3 þ(NH4)2SO4
(22)
Even the all salts are dissolved in the reaction mixture, due to the high association constant of SBC (0.72) and SC (8.3) [33] makes remain intact and makes easy to differentiate the carbonate and bicarbonate via SBC and SC and the obtained results are depicted in Fig.4. The AC both þve and –ve error% is controlled within 75% in high end ratios. It is believed that the Na þ ion effectively replaces the NH4þ of AC. In the case of ABC determinations, the excess Na2SO4 shows controlled error percentages at most ratios 0.6 Na2SO4 presence
Found difference(g)
0.4 0.2 0.0 -0.2
(NH4)2CO3
-0.4
NH4HCO3
-0.6
0.6 (NH4)2SO4 presence
0.4
Found difference (g)
(found value is less than originally taken). The (NH4)2CO3 error was a minimum of 5.2 þ ve% when carbonate to bicarbonate ratio is 2:8 and with increasing AC ratio –ve error% increases to 410.0%. Except titration using phosphoric acid, the other two acids CH3COOH (plot (b) in Fig.3) and HCl (plot (c) in Fig.3) shows a similar trend and vice versa for ABC. The titration of (NH4)2CO3/ (NH4)HCO3 mixture with phosphoric acid (plot (a) in Fig.3) shows ABC of 8.7 þ ve error% when carbonate to bicarbonate ratio is 2:8, which increases with decreasing of ABC ratio. In a similar way, AC shows –ve error% at lower ratio that increases with decreasing AC, as shown in Fig.3(plot (a)). Among three acids, titration using HCl shows (Fig.3(plot (c)) –ve error% in both AC and ABC differentiation with the existence of a similar equilibrium error problem as found in back titration method. The error% of the both AC and ABC might have also from ACm formation in the first step using phenolpthalein titration where ACm converted to CAm directly, instead of ABC formation, as seen in equation 18’. This ABC and ACm equilibrium (eqn. 19’) at high end ratios of AC and ABC might have influenced to receive this high error% in the both –ve and þve. In higher ABC, more of ACm might have formed and vice versa at lower ABC that makes an equilibrium variation depending upon the concentration of ABC, which leads high error% in their determination. Further works can be carried out using HCl acid.
331
0.2 0.0 -0.2
(NH4)2CO3 NH4HCO3
-0.4 -0.6 -0.8 2:8
4:6
5:5
6:4
7:3
8:2
9:1
Concn. by ratio Fig. 5. Effect of (NH4)2SO4 with different ratios of (NH4)2CO3/(NH4)HCO3 mixture on their identification by double indicator method. Added (NH4)2SO4 concentration¼ 0.15 M.
except high end ratio of 9:1, indicating effective replacement of the Na þ ion in ABC. The reason why the minimized error occurred is, only a small amount of AC becomes ACm that might not converted to SC while addition of SS due to sodium ammonium carbamate and double salt formation as shown in eq. 23. But one positive effect of the presence of sodium sulfate is that it enables the differentiation of ABC and AC without its interference. Nonetheless, more accurate differentiation is required regardless of ABC/AC ratio. Na2SO4 þNH2COONH4-NH2COONaþNH4.Na.SO4
(23)
3.4. Approach through the effect of equilibrium control by addition of salt and temperature As explained by eqs. 1 and 2, a common ion in the equilibrium change is sodium ion. One possibility of minimizing equilibrium change involves the addition of excess ammonium ion to the titration solution, here ammonium sulfate is probably a good choice, as shown by eq. 25. Ammonium sulfate added titration results to identify the ABC and AC are depicted in Fig. 5. There appears the error% for both þve and –ve errors has controlled within in 3 71% upto the ratios of 2:8 to 6:4 of AC and ABC. Also, there might be an equilibrium arrest between AC and ACm if add the AS (eqn. 24) for this both þve and –ve error% minimization. Beyond these ratios, the both -ve and þve error% were not controlled within the error limit. These results explain higher the AC concentration in the reaction mixture makes higher the both –ve and þve error%, which means the equilibrium only between AC and ABC at high concentration. In other words, at high ratios of AC, AS is not able to control the conversion to ABC, as shown in eq. 25. But, the both –ve and þve error% are almost controlled in all the ratios of AC and ABC when compared with no salt addition and in presence of Na2SO4, as shown in Fig.7. (NH4)2SO4 (24) NH2COONH4 (NH4)2CO3 At higher AC ratio:
-0.8 2:8
4:6
5:5
6:4
7:3
8:1
9:1
Concn. by ratio Fig. 4. Replacement titration with Na2SO4 on different ratios of (NH4)2CO3/(NH4) HCO3 mixture on their differentiation by double indicator method. Added Na2SO4 concentration¼0.14 M.
(NH4)2CO3(n) þH2Oþ NH4HCO3(n)-(NH4)2CO3(n-1) þ(NH4)HCO3 (n þ 1) þNH4OH(n)
(25)
It is seen from eq. 13 that there is a possibility of AC decomposition to ABC via H2O at high concentration ratios and forms NH4OH in
332
M. Govindan et al. / Talanta 132 (2015) 327–333
0.6
ACm
AS
AC
0.4
AS 5 °C
ABC
Found difference (g)
Scheme 1. Equilibrium arrest between AC, ACm, and ABC at different conditions.
0.2 0.0 -0.2 -0.4 -0.6 -0.8 25
20
15
10
5
ο
Temperature ( C) Fig. 6. Effect of temperature with high ratios of (NH4)2CO3/(NH4)HCO3 mixture on their differentiation in presence of (NH4)2SO4 by double indicator method. Added (NH4)2SO4 concentration¼ 0.15 M. Insert figure: Concentration difference effect in absence of (NH4)2SO4 at 5 1C on different ratios of (NH4)2CO3/(NH4)HCO3.
increasing AC to ABC ratios, the differentiation error% is increased up to 20%, which indicates AS is the only component responsible to arrest or control the equilibrium transfer between AC to ACm, and AC to ABC, as shown in Scheme 1. The both AS presence (AC and ACm equilibrium arrested) and low temperature (AC and ABC equilibrium arrested) give highly synergistic effect for clear differentiation of AC and ABC at all ratios. For clear differentiation, comparative results are plotted and presented in Fig. 7, here shown how the error trend of AC and ABC is minimized at different experimental conditions. It clearly explains the accurate identification ABC and AC is only possible by titration in the presence of AS at 5 1C, especially along with impurities. We believe that this present condition may support to get more accurate identification when analyze through an instrument. At 5 1C in presence of AS in all ratios: 3.5. Validation through real sample
Fig. 7. Overall comparitive plot on differentiation of ((NH4)2CO3/(NH4)HCO3 ratios with different experimental conditions using double indicator method: (a) without addition of salts; (b) With Na2SO4; (c) With (NH4)2SO4; (d) At 5 1C with (NH4)2SO4.
solution or NH3 might have liberated from solution. Moreover, the concentration we took is very less (0.1 M) for titration, which makes no loss of NH3 by liberation. We tried to check the pH value after phenolpthalein titration, but found no change or high pH value. It is well known that the reaction can be arrested at low temperature [34], and thus, there is a possibility that the conversion of AC could be controlled by decreasing temperature. As a first trial, the higher AC ratios were considered to check temperature effect and the obtained results are presented in Fig. 6. The five different temperatures examined from 25 to 5 1C shows a clear decrease in the both –ve and þve error%. Irrespective of the higher three ratios, the both –ve and þve error% decreased with decreasing temperature to within an error% of (3 71). Nonetheless, only low temperature (5 1C) with absence of AS shows ca. 20% error in both þve and –ve in all the ratios, as shown in the insert figure of Fig.6. At lower ratios. 2:8 and 4:6, the ABC differentiation is very effective (within 3% error) at 5 1C in the absence of AS and it starts increase with
The desulfurized crude sample was first purified by adding water and then reacted with NH3 and CO2 to produce NaHCO3. In order to finalize the reaction conditions, ABC and AC determination are important in a secondary waste solution of AS, as shown in reaction 26. The total reaction process was carried out as shown in Fig.1. In brief, 11.1 Kg of crude desulfurised waste were dissolved in 24.12 L of water. The slurry solution heated to 451C along with stirring about 30 min. Then the slurry was then transferred to filter press machine, which was operated at 451C and 4 atm pressure. Typically 11 wt% of insoluble impurities were collected from the filter press. The filtered solution has analyzed by titration for carbonate and bicarbonate, ISE for sodium ion, IC for both cations and anions. In these given conditions, approximately 90 wt% (obtained salt weight/(total raw waste weight-impurities weight) x 100) sodium sources through filter press method have been recovered. The exact weight and volume that obtained were depicted in row 7 in Table 1. The analyzed filtered desulfurized solution was then sent to carbonation tank to produce NaHCO3 using ammonia solution and CO2 gas. Based on the analysis value of sodium source, the ammonia solution 0.9 mol ratio against sodium source was taken for the production of sodium bicarbonate. Under optimized conditions of temperature, water volume, and pressure, sodium bicarbonate production process were done using the carbonation tank. Then the reaction slurry sent to peeler type decanter modified with the overhead temperature controller. The peeler type decanter was operated at 45 1C temperatures both on overhead and reactants slurry, 800 mesh filter and 3500 rpm. The product error was predicted based on the analyses of both solid and secondary waste liquid by titration, UV-Visible spectroscopy, and ISE methods. The total ions of both solid and liquid that found are displayed in row 11–12 in Table 1. The found difference of BC and C in the presence of all is 27 0.5%. This indicates validation of the present method and very crucial to finalize the further modification in perfection process. Crude wasteþNH3 þCO2 þH2O-NaHCO3 þ (NH4)2SO4
(26)
4. Conclusions The AC and ABC were identified within an error% successfully by simple titration. Back titration results confirm it will differentiate
M. Govindan et al. / Talanta 132 (2015) 327–333
333
Table 1 Analysis results from regeneration reaction process of NaHCO3 in desulfurized waste. Raw sample weight(Kg) Water volume (L) Separation temperature Impurity weight (Kg) and % Filtrate soln. obtained (L) Sodium sources wt (Kg) and wt%
For preparation of NaHCO3 Water volume for NaHCO3 production (L) Reaction tempr. (1C) and Time (min) Needed CO2 (Kg) ammonia (Kg) and vol. (L) Product weight (Kg) and wt% Analysis results Ions name SO24 CO23 HCO3 NH4þ Na þ
11.1 24.12 45 1C 1.21; 10.92% 23.12 Na2SO4 ¼ 7.66 Na2CO3 ¼ 0.99 NaHCO3 ¼0.33 Total weight¼ 8.98 Yield in %¼ 8.98/9.89(11.1-1.21¼ 9.89) ¼90.78 23.12 35–38 and 60 min NH3 (4.519 Kg) or NH4OH solution (8.319 L): Need CO2 (11.677 Kg) 8.34; 84.33(8.34/9.89) Reactants weight (Kg) Found weight in solid (Kg) Found weight in solution (Kg) 5.178 0.119 4.904 0.559 0.018 0.531 6.157 5.585 0.559 2.324 0.022 2.161 2.999 2.177 0.805
only one ratio of AC and ABC (6:4) and leave a high error to all other ratios makes unsuitable in their complete identification. Also, the double indicator titration method confirms, similar as the back titration method, not suitable to identify the AC and ABC except one ratio (5:5). In a replacement titration through the addition of Na2SO4 minimizes the identification of AC and ABC error% a less than the only double indicator method confirms partial equilibrium arrest between ACm and SC and conversion of AC and ABC to SC and SBC. Instead of SS, AS addition give 37 1 error% up to the ratios of 6:4 confirms the effective minimization in complete equilibrium arrest between ACm and AC, and conversion of AC to ABC. Beyond this 6:4 ratio, the error% has increased with increase in ABC concentration, which concludes only equilibrium between AC and ABC conversion. Finally, at low temperature 5 1C, the error% is completely arrested all equilibriums and minimized the error%, even at high ratios 7:3 of AC and ABC. The validation of the present method in real sample analysis is perfectly working to identify the BC and C in the presence of excess AS within an allowed error%. These results conclude that low temperature in the presence of excess AS is more suitable to differentiate the AC and ABC by titration analysis. Acknowledgements This study was supported by the Korea Research Foundation and the Korean Federation of Science and Technology Societies funded by Korea Government (MOEHRD, Basic Research Promotion Fund) and the National Research Foundation of Korea (NRF) funded by the Korean government (MEST) (Grant Nos. 20100027330 & 2014001974). Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.talanta.2014.08.044. References [1] [2] [3] [4]
I. Dahlan, K.T. Lee, A.R. Mohamed, J. Sci. Ind. Res 163 (2013) 83–87. R.-B. Lin, S.-M. Shih, C.-F. Liu, Chem. Eng. Sci. 58 (2003) 3659–3668. C.F. Liu, S.-M. Shih, Ind. Eng. Chem. Res. 45 (2006) 8765–8769. C. Wu, S.-J. Khang, T.C. Keener, S.-K. Lee, Adv. Environ. Res. 8 (2004) 655–666.
% error 3.08 1.85 2.28 5.24 1.83
[5] L. Xu, J. Guo, F. Jin, H. Zeng, Chemosphere 62 (2006) 823–826. [6] R. Sundararaman, C. Song, Energ. Fuel. 27 (2013) 6372–6376. [7] H. Deng, H. Yi, X. Tang, H. Liu, X. Zhou, Ind. Eng. Chem. Res. 52 (2013) 6778–6784. [8] I. Dahlan, K.T. Lee, A.H. Kamaruddin, A.R. Mohamed, Environ. Sci. Technol. 40 (2006) 6032–6037. [9] G. Muthuraman, S.J. Chung, I.S. Moon, J. Ind. Eng. Chem. 51 (2012) 2697–2703. [10] K.C. Pillai, S.J. Chung, T. Raju, I.S. Moon, Chemosphere 76 (2009) 657–664. [11] J. Bichel, S. Stephen, Method for recovering purified sodium bicarbonate and ammonium sulfate. U.S. Patent, 0156775 A1, Aug. 12, 2004. [12] F. Christensson, H.C.S. Koeford, A.C. Petersen, K. Rasmussen, Acta Chem. Scand. A 32 (1978) 15–17. [13] W.C. Femelius, in: W.C. Femelius (Ed.), Inorganic Syntheses, Vol. II, McGrawHill Book Company Inc, New York, 1946, pp. 85–86 (Chapter 4). [14] J.T. Yeh, K.P. Resnik, K. Rygle, H.W. Pennline, Fuel Process. Technol. 86 (2005) 1533–1546. [15] J.F.I. Thomas, J.J. Lynch, J. Amer. Water Works Assoc 52 (1960) 259. [16] A.A.B. Pichle, M. Cefola, Ind. Eng. Chem 6 (1939) 327–332. [17] A. Suchier, Angew. Chem. 44 (1931) 534–536. [18] R Belcher, A.J. Nutten, A.M.G. Macdonald, Quantitative Inorganic Analysis, Third ed., Butterworths, London, 1970. [19] C.L. Wilson, D.W. Wilson, Comprehensive Analytical Chemistry, Elsevier, London, 1960 (Classical Analysis). [20] N. Wen, M.H. Brooker, J. Phys. Chem. 99 (1995) 359–368. [21] L. Meng, S. Burris, H. Bui, W.P. Pan, Anal, Chem 77 (2007) 5947–5952. [22] Q. Zhao, S. Wang, F Qin, C. Chen, Ind. Eng. Chem. Res. 50 (2011) 5316–5325. [23] F. Mani, M. Peruzzini, P. Stoppioni, Green Chem. 8 (2006) 995–1000. [24] R. Bosco, B. Grandjean, Using raman spectroscppy to control carbonate/ bicarbonate concentrations, U.S. 2011, 0125326 A1, May 26, 2011. [25] P.L. Fosbøl, K. Thomsen, E.H. Stenby, Ind. Eng. Chem. Res. 48 (2009) 4565– 4578. [26] M. Laliberte, J. Chem. Eng. Data. 52 (2007) 321–335. [27] H. Huang, S.G. Chang, Energ. Fuel. 16 (2002) 904–910. [28] W.T. Bolleter, C.J. Bushman, P.W. Tidwell, Anal. Chem. 33 (1961) 592–594. [29] S.A. Ben Mussa, H.S. Elferjani, F.A. Haroun, F.F. Abdelnabi, Int. J. Pharm.Tech. Res 1 (2009) 598–604. [30] F. Qin, S. Wang, I. Kim, H.F. Svendsen, C. Chen, Int. J. Green. Gas Con 5 (2011) 405–412. [31] G.G. Wildgoose, A.T. Masheter, A. Crossley, J.H. Jones, R.G. Compton, Int. J. Electrochem. Sci. 2 (2007) 809–819. [32] M. Gronvald, C. Faurhold, Acta Chem. Scand. 14 (1960) 1374–1380. [33] P.D. Wimberley, O. Siggaard-Anderson, N. Fogh-Anderson, A.B.T. Boink, Scand. J. Lab. Inv. 45 (1985) 7–10. [34] G. Beck, J. Janek, Solid State Ionics 170 (2004) 129–133.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Effective identification of (NH4)2CO3 and NH4HCO3 concentrations in NaHCO3 regeneration process from desulfurized waste Muthuraman Govindan 1, Kannan Karunakaran 1, Palanisami Nallasamy 1, Il Shik Moon n Department of Chemical Engineering, Sunchon National University, 315 Maegok Dong, Suncheon 540-742, Chonnam, South Korea
art ic l e i nf o
a b s t r a c t
Article history: Received 6 June 2014 Received in revised form 13 August 2014 Accepted 14 August 2014 Available online 28 August 2014
This work describes the quantitative analysis of (NH4)2CO3 and NH4HCO3 using a simple solution phase titration method. Back titration results at various (NH4)2CO3–NH4HCO3 ratios demonstrated that 6:4 ratio caused a 3% error in their differentiation, but very high errors were found at other ratios. A similar trend was observed for the double indicator method, especially when strong acid HCl was used as a titrant, where still less errors (2.5%) at a middle ratio of (NH4)2CO3–NH4HCO3 was found. Remaining ratios with low (NH4)2CO3 (2:8, 4:6) show high þve error (found concentration is less) and high (NH4)2CO3 (7:3, 8:2, and 9:1) show high –ve error (found concentration is higher) and vice versa for NH4HCO3. In replacement titration using Na2SO4, at both higher end ratios of (NH4)2CO3–NH4HCO3 (2:8 and 9:1), both –ve and þve errors were minimized to 75% by partial equilibrium arrest between (NH4)2CO3 and NH2COONH4, instead of more than 100% observed in back titration and only double indicator methods. In the presence of (NH4)2SO4 both –ve and þve error% are completely reduced to 3 7 1 at ratios 2:8, 4:6, and 6:4 of (NH4)2CO3–NH4HCO3, which demonstrates that the equilibrium transformation between NH2COONH4 and (NH4)2CO3 is completely controlled. The titration conducted at lower temperature (5 1C) in the presence of (NH4)2SO4 at higher ratios of (NH4)2CO3–NH4HCO3 (7:3, 8:2, and 9:1) shows complete minimization of both –ve and þve errors to 2 71%, which explains the complete arresting of equilibrium transformation. Finally, the developed method shows 27 1% error in differentiation of CO23 and HCO3 in the regeneration process of NaHCO3 from crude desulfurized sample. The developed method is more promising to differentiate CO23 and HCO3 in industrial applications. & 2014 Elsevier B.V. All rights reserved.
Keywords: Flue Gas Desulfurization Sodium sulfate Ammonium sulfate Ammonium carbonate Ammonium bicarbonate Titration
1. Introduction Growing awareness of the risks of climate change has generated public concerns, and over the last two decades the sequestering of air pollutants has been of major research interest. As the world's population and energy demand increase, continued industrialization will undoubtedly increase levels of atmospheric pollutants like NOx, SOx, and CO2. In particular power plant and steel plant furnaces emit SOx and NOx (called flue gases), which are removed using wet or dry sorbent methods [1]. Flue Gas Desulfurization (FGD) is performed using the wet sorbent method. However, due to the large amount of space required for installation, the large volume of water required and high capital and operating costs [2], research has shifted toward dry methods of SO2 removal. Various types of solid sorbent/catalysts are being used for dry FGD, such as, calcium based [3], sodium based [4], activated carbon [5],
n
Corresponding author. Tel.: þ 82 61 750 3581; fax: þ 82 61 750 3581. E-mail address: [email protected] (I.S. Moon). 1 These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.talanta.2014.08.044 0039-9140/& 2014 Elsevier B.V. All rights reserved.
metal oxide [6], and zeolite sorbents [7], waste-derived siliceous materials [8], and mediated electrocatalytic oxidation (MEO) [9,10]. Many of these methods have limitations in terms of their commercialization. Among these methods, sodium-based sorbents are considered more effective for FGD, especially for SOx. These processes are conducted mainly using sodium carbonate (Na2CO3), sodium bicarbonate (NaHCO3), and trona (Na2CO3 NaHCO3 2H2O). In this process, either sodium bicarbonate (NaHCO3) (or) sodium carbonate (Na2CO3) reacts with SO2 to form Na2SO3 which is further converted to sodium sulfate (Na2SO4) in the presence of excess oxygen, as explained by the following equation sequences [1–4]: 2NaHCO3 þSO2-Na2SO3 þ 2CO2 þH2O
(1)
2NaHCO3-Na2CO3 þCO2 þH2O
(2)
Na2CO3 þSO2-Na2SO3 þ CO2
(3)
Na2SO3 þ1/2O2-Na2SO4
(4)
However, it had a few drawbacks: (i) sodium bicarbonate is expensive; (ii) the resulting by-product from the sodium
328
M. Govindan et al. / Talanta 132 (2015) 327–333
NH3storage
Waste solution
CO2storage
Gas
Homogenizer with controlled temperature
Crude desulfurized waste mixture
Separator
Ammonium sulfate secondary waste
Reactor with controlled temperature
L S
L S Applications
Impurities
NaHCO3
Dryer
Fig. 1. Schematic representation for the regeneration pathway of NaHCO3 from desulfurized waste sample.
bicarbonate–SOx reaction (sodium sulfate) is of limited economic value; and (iii) sodium sulfate disposal is expensive and poses significant environmental challenges. To regenerate sodium bicarbonate from secondary waste Na2SO4 (SS) provides an efficient way of avoiding the above mentioned drawbacks [11]. The regeneration of sodium bicarbonate from sodium sulfate is schematically presented in Fig. 1. The regeneration of sodium bicarbonate involves three steps: (i) separation of the sodium sources from their impurities, (ii) regeneration of NaHCO3 (SBC) using NH4HCO3 (ABC) or gases of NH3 and CO2, and (iii) utilization of secondary waste solution (NH4)2SO4 (AS). In these regeneration processes, analysis of each ion plays a crucial role in finalizing the experimental conditions such as the molar ratio of reactants and water content. Particularly the secondary waste AS containg solution, which includes additionally (NH4)2CO3 (AC) and ABC, should be exactly analyzed to perfect the minimization of error. Identifications of CO23 and HCO3 ions are very difficult through analysis by titration due to their complex equilibrium process as follows [5–9,12,13]: þ þ NH4(aq) þ H2O(l)⇌NH3(aq) þH3O(aq)
(5)
þH2O CO23(aq)
(6)
(aq)⇌HCO3(aq) þOH(aq)
þNH4þ (aq)⇌HCO3(aq) þNH3(aq) CO23(aq)
(7)
þNH3(aq)⇌H2NCOO(aq) þH2O(l) HCO3(aq)
(8)
þ þ H2O(l)⇌CO23(aq) þNH4(aq) H2NCOO(aq)
CO23
(9) HCO3
and by a back titration First, the differentiation of method using BaCl2 as a precipitant for CaCO3 has been identified to be infeasible [14]. They found more restriction in minimizing the error in quantification of each compound and no detailed studies about various concentration ratios of CO23 and HCO3 [15]. Later, double indicator Warder's titration method was used to differentiate the CO23 and HCO3 as a next level [16,17]. However, quantifying the CO23 and HCO3 is difficult due to lack of endpoint identification; color change is very feeble in case of AC and ABC [18,19], and no detailed study has been performed by the titration method. In the late 1990s, it was reported that FT-IR, NIR, Raman spectroscopies and powder-XRD can differentiate AC and ABC quantitatively and qualitatively [20–22]. The XRD 2θ peaks of ABC and AC are at 16.5 and 34.4, respectively, and can be used to analyze the solid state compositions of mixtures [21]. In the case of
FT-IR analysis, there is a possibility of quantitative differentiation of ABC (1200 1450 cm 1 for asymmetric CO stretch and 1600 cm 1 for symmetric CO stretch) and AC ( 1300 1600 cm 1 for asymmetric CO stretch). The FT-IR spectra of both solid and solution phases of AC and ABC are quite similar and difficult to distinguish because of broad peaks due to both carbonate and bicarbonate between 1200 and 1700 cm 1 and the same peak at 3098 cm 1, which is due to the ammonium ion [22]. As reported by Meng et al., by variation of oxygen hydrogen (OH) bond of ABC at 1450 nm, which is not present in the carbonate, one can differentiate quantitatively in solid state samples [21]. However, it is difficult to distinguish the bicarbonate from carbonate ion in aqueous solution due to interference of hydroxyl (OH) groups in the water, which restricts the NIR analysis in solution samples. 13C NMR studies have demonstrated the qualitative differentiation of ABC and AC at room temperature [23]. Even the fast equilibrium of HCO3 /CO23 anions results in their two NMR peaks, which permits identifying the HCO3 /CO23 ions quantitatively. Through the Raman spectra, the ABC and AC in solution phase have been differentiated quantitatively using scattered light of the peaks at 1015 cm 1 (for bicarbonate) and 1065 cm 1 (for carbonate) [24]. However, Raman analysis needs an internal standard for exact differentiation of both ABC and AC. The mixture of our secondary waste contains internal standard compound Na2SO4, which should be analyzed additionally using another method. Accordingly several issues must be resolved for the quantitative analyses of ABC and AC in the solution phase, especially in samples obtained during the regeneration of SBC from desulfurized crude waste. There are some models to control the reaction and its coordinates to get a high yield and purity in regeneration of SBC from desulfurized waste [25,26]. Currently there are no reports on controlling the reaction based on analysis of ions present in the crude waste (before reaction) and secondary product waste (after reaction). In the present work a method is developed based on analysis of ions present in crude waste and secondary waste, to control the reaction in the SBC regeneration process. Herein, validations of back titration at different ratios of AC ad ABC were performed for its confirmation. Then, the double indicator method was repeated with different ratios of AC and ABC to check its applicability for their differentiation. Additionally replacement titration, by SBC, was studied. In order to check the equilibrium of CO23 and HCO3 , AS and temperature effect were studied.
M. Govindan et al. / Talanta 132 (2015) 327–333
Finally, the developed method was tested on a real sample from desulfurized crude waste regeneration process of the SBC.
2. Experimental section 2.1. Chemicals Ammonium bicarbonate and ammonium carbonates were purchased from Aldrich. Phenolphthalein, methyl orange, bromocreasol green, phenol red, sodium carbonate, barium chloride dehydrate, sodium hydroxide, and phosphoric acid were purchased from Daejung Chemicals, Korea. Hydrochloric acid and acetic acid were purchased from OCI, and Duksan Korea respectively. 2.2. Reagents 0.1 M stock solutions of hydrochloric acid, acetic acid and phosphoric acid were prepared by dissolving approximately 8.4 ml of 11.6 M hydrochloric acid in 1 L of deionized water, 5.8 ml of 17.4 M acetic acid in 1 L of deionized water, and 6.8 mL of 14.6 M phosphoric acid in 1 L of deionized water respectively. These solutions were standardized by titration against standard sodium carbonate. 2.3. Back titration This method is the oldest method to differentiate two components by precipitating one component [14]. In order to differentiate AC and ABC, various ratios such as (2:8), (4:6), (5:5), (6:4), (7:3), (8:2), (9:1) were chosen. The back titration was performed using two steps. In the first step, total alkalinity was determined by placing a 5 ml sample in a 100 ml conical flask, adding methyl orange indicator and titrating against standardized 0.1 M HCl, as shown in Eq. (10). The end point change from yellow to red was considered as a measure of total alkalinity. This procedure was repeated till concordant values were obtained and the average value was taken. In the second step, 5 ml of sample was taken in a 100 ml conical flask, then 16 ml of NaOH and 0.88 ml of 10% BaCl2 solution were added along with a few drops of phenolphthalein indicator. Instantly, the white precipitate (BaCO3) generated in the sample mixture solution was then immediately titrated against standardized 0.1 M HCl solution after using a high speed centrifuge for a short time (30 s), as shown in Eq. 11. The disappearance of pink color is an indication of end point. The volume of consumed HCl was recorded and the test was repeated for concordant values. Step 1: Total alkalinity: NH2COONH4 and (NH4)2CO3 þ NH4HCO3 þ3HCl-3NH4 Cl þ2H2Oþ2CO2
(10)
329
The same method has been adopted to identify ammonium carbonate and ammonium bicarbonate [27]. Briefly, ammonium carbonate and ammonium bicarbonate mixture was dissolved in 100 ml of water. 5 ml of this solution was transferred into a conical flask. 2 drops of phenolphthalein were added and pink color appeared. When titrated against 0.1 M acids the pink color disappeared, which is the endpoint of ammonium carbonate solution. To the same solution 2 drops of phenol red were added and the solution became yellow. After titration again with acetic acid the yellow turned into red, which is the ammonium bicarbonate endpoint. For phosphoric acid and hydrochloric acid, methyl orange is used for ammonium bicarbonate and the end point is orange to red. In all titration experiments, the titration time was 0.5–3 min for all titrations at all ratios of AC and ABC. In the AC determination step, the time spent is between 0.5 and 1.5 min. The error% of ABC was calculated by substracting the AC value obtained from the phenolpthalein indicator by total alkalinity (ACl) value obtained from the second indicator (methyl orange). As shown in Eqs. (12)–(14). to cross check, the AC value added with ABC, which is equal to carbamic acid (CAm) or carbonic acid (CA) that developed during titration. Then, the CAm or CA was subtracted by ACl to get right ABC. CACl – CAC or CACm ¼ CABC
(12)
CAC þ CABC ¼CCAm
(13)
CACl – CCAm
or CA
or CA ¼CABC
(14)
In the replacement titration, two times the concentration of Na2SO4 was added and repeated as a double indicator method of titration. In stabilization experiments, (NH4)2SO4 and temperature were included as two times higher molar concentration and 25 to 5 1C variation respectively. The room temperature experiments were carried out at 207 2 1C. All the experiments were repeated until the results were reproducible. 2.5. Analysis of real regeneration process samples First, an industrial desulfurized sample 10 g was dissolved in 250 ml of deionized water and filtered to remove impurities, and separated the salt solution for analysis. The Na þ and NH4þ ions were estimated using ion selective electrodes (ISE) that purchased from VAN LONDON Phoenix.CO (No.: NA715XX for Naþ and No.: NH415XX) with the help of iSTEK multimeter (Model No.: pH-240 L) from USA. Prior to the analysis, the electrodes and instruments were calibrated as per the instruction that given by a manufacturing company. In another way, the NH4þ was cross checked with standard procedure by UV-Visible spectroscopy (Shimadzu, UV160A) from Japan, by the indophenol blue method at 630 nm [28]. The sulphate ion in the sample was estimated in the form of Barium sulphate by the addition of Barium chloride precipitating agent along with Conditional reagent followed by UV–vis spectrophotometric absorption at 420 nm [29]. All metal ions that present in the real sample was analyzed by ICP-MS (model ELANDRCe, Perkin Elmer) from USA.
Step 2: Differentiate HCO3 and CO23 2HCl
ðNH4Þ2 CO3 þ NH4 HCO3 þ NaOH þBaCl2 ⟹3NH4 Cl þ BaCO3 þ H2 O þ NaCl þ H2 CO3 ð11Þ
2.4. Double indicator method There are some reports for differentiation of sodium carbonate and bicarbonate in the reaction mixture [16,17], and a few reports for ABC and AC without detailed experimental conditions [15].
3. Result and discussion 3.1. Conventional back titration on differentiation of (NH4)2CO3/ (NH4)HCO3 As evidenced in previous reports, SBC and SC can be differentiated effectively by back titration with an error of less than 5%. However, they never considered ACm equilibrium in back titration to differentiate ABC and AC in CO2 absorption process using ammonia [14]. In the present case, desulfurized waste contains many salts that including SBC and SC in different ratios, which
330
M. Govindan et al. / Talanta 132 (2015) 327–333
3.2. Double indicator method (NH4)2CO3 NH4HCO3
Found difference (g)
0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8
2:8
4:6
5:5
6:4
7:3
8:2
9:1
Concn. by ratio Fig. 2. Effect of different concentration ratios on found concentration differences of (NH4)2CO3 and (NH4)HCO3 in their identification through back titration.
must be accurately identified for further reaction with ABC to regenerate SBC. After reaction or to control the reaction, differentiation of ABC and AC is important. By keeping all in mind, we have started optimizing conditions in back titration such as the effect of NaOH and BaCl2 addition time in particular of AC to ABC ratio 6:4. The obtained results are presented in supporting information Table S1. There appears minimized error% during the both NaOH and BaCl2 addition time in 30 s. Using these selected conditions, the back titrations to identify the ABC and AC in different ratios at room temperature was performed and the obtained results are depicted in Fig.2. At high concentration of AC, the found concentration of AC increased to 0.35 g than (þve error) the actual input value. At the same time, the found ABC is very decreased to 0.54 g (-ve error) than the originally added in 8:2 ratio. The found difference of both ABC and AC increased and decreased in a similar manner with increasing ratio up to 6:4, where the found difference is almost 0.01 g which equal to less than 3% þve error. Further increasing the ratios beyond 6:4 of AC and ABC, the found difference shows a reverse trend which means, the found AC is decreased than the originally added and ABC is increased than the input concentrations. The overall found difference explains that the highest concentration of AC or ABC are tended to convert to another form chemically. The chemical equilibrium may dominate more at high end concentration ratios. It is reported that the ABC is converted to AC at high temperature and excess of CO2 [27]. In a similar way, AC converted to ABC in excess CO2 and H2O [30]. But only containing of both ABC and AC, how the conversion has occurred? Instead of conversion, we suggest the possibility of equilibrium stabilization, as shown in the following eqs. (15 and 16). Another possible way for high error % is the formation of ACm from AC in water as shown in eqs. 8 and 9. It is evidenced that the ACm will not precipitate in precipitation step upon addition of BaCl2 [31] to remove the AC from the titer solution in related to avoid its interference during the back titration of the consumed NaOH with HCl. In case ACm remains in titer solution, the consumed HCl volume will vary which in turn leads high error% in determination of AC, as shown in eq. (17) At High (NH4)2CO3: (NH4)2CO3 "NH4HCO3
(15)
At high NH4HCO3: (NH4)2CO3" NH4HCO3
(16)
(NH4)2CO3 þ NH4HCO3 þHCl-2NH4HCO3 þNH4Cl
(18)
NH2COONH4 þNH4HCO3 þHCl-NH4HCO3 þNH2COOHþNH4Cl (180 ) 2NH4HCO3 þ2HCl-2NH4Cl þ2H2CO3
(19)
NH4 HCO3 þ NH2 COOH þ HCl!NH2 COOH þ H2 CO3 þ NH4 Cl
ð190 Þ
NH2COOHþHCl-NH4ClþCO2
(200 )
2
Depending upon the acid, which we used, the error% is varying for both the ABC and AC or ACm that can be defined as the -ve error% (found value is higher than originally taken) and þ ve error% 0.4 0.0 -0.4 -0.8
(NH4)2CO3 NH4HCO3
0.4 0.0 -0.4 -0.8 0.4 0.0 -0.4 -0.8 2:8
4:6
5:5
6:4
7:3
8:2
9:1
Concn, by ratio
HCL
NH2 COONH4 þ NH4 HCO3 þ NaOH þ BaCl2 !2NH4 Cl þ BaCO3 þ H2 O þ NaCl þ NH2 COOH
Because of defined differences in the pH value of ABC (8.3), and AC (9), the two indicator method could possibly provide accurate ABC and AC concentrations. In consideration of ACm, the pH of ACm 8.9 [32], which is very close to the pH of AC, it clearly helps to differentiate ABC even if it exists because of phenolpthalein used as indicator. Since ABC, ACm, and AC are weak bases, acid selection could minimize errors. We selected three different acids, that is, acetic acid (weak base vs weak acid) and phosphoric and hydrochloric acids (weak base vs strong acid). As evident in literature, In order to know the time dependent equilibrium of AC, ACm, and ABC, the sample titration time at minimum (1 min) and maximum (720 min) were done and the results are tabulated and presented in supporting information Table S2. After 1 min, the error% has increased to only 7 to 10 at AC to ABC of 75:25 ratio. In consideration of practical difficulties, 1 min sampling time was selected for all the titrations. During analysis, we found difficulties in end point measurements, especially for AC and ACm determinations, when only a pale pink was obtained after adding phenolphthalein. This color change behavior just contrary to the SBC and SC differentiation, there was able to judge a clear difference in end point and correct value. The titration results obtained for ABC and AC are depicted in Fig. 3. In plot (b) of Fig.3 illustrates the titration of acetic acid with (NH4)2CO3/(NH4)HCO3 mixture. During titration, AC or ACm becomes ABC or CAm and carbonic acid (CA) at second endpoint while the use of methyl orange indicator as shown in eqs. 18–20’ in case of ACm. The % errors were calculated using eqs. (12–14).
Found difference (g)
0.6
ð17Þ
Fig. 3. Effect of acids on titration of different concentration ratios of (NH4)2CO3/ (NH4)HCO3 mixture in their identification by double indicator method: (a) Phosphoric acid; (b) Acetic acid; (c) Hydrochloric acid.
M. Govindan et al. / Talanta 132 (2015) 327–333
3.3. Identification through replacement titration by Na2SO4 It is known that alkali salts reacts with ABC and AC form alkali carbonate and bicarbonate [11]. Using this approach, sodium sulfate was added accordingly to the concentrations of ABC and AC to convert them to SBC and SC, as shown in the eqs. 21, and 22, this facilitates the accurate differentiation of ABC and AC. Na2SO4 þ(NH4)2CO3-Na2CO3 þ(NH4)2SO4
(21)
Na2SO4 þ2(NH4)HCO3-2NaHCO3 þ(NH4)2SO4
(22)
Even the all salts are dissolved in the reaction mixture, due to the high association constant of SBC (0.72) and SC (8.3) [33] makes remain intact and makes easy to differentiate the carbonate and bicarbonate via SBC and SC and the obtained results are depicted in Fig.4. The AC both þve and –ve error% is controlled within 75% in high end ratios. It is believed that the Na þ ion effectively replaces the NH4þ of AC. In the case of ABC determinations, the excess Na2SO4 shows controlled error percentages at most ratios 0.6 Na2SO4 presence
Found difference(g)
0.4 0.2 0.0 -0.2
(NH4)2CO3
-0.4
NH4HCO3
-0.6
0.6 (NH4)2SO4 presence
0.4
Found difference (g)
(found value is less than originally taken). The (NH4)2CO3 error was a minimum of 5.2 þ ve% when carbonate to bicarbonate ratio is 2:8 and with increasing AC ratio –ve error% increases to 410.0%. Except titration using phosphoric acid, the other two acids CH3COOH (plot (b) in Fig.3) and HCl (plot (c) in Fig.3) shows a similar trend and vice versa for ABC. The titration of (NH4)2CO3/ (NH4)HCO3 mixture with phosphoric acid (plot (a) in Fig.3) shows ABC of 8.7 þ ve error% when carbonate to bicarbonate ratio is 2:8, which increases with decreasing of ABC ratio. In a similar way, AC shows –ve error% at lower ratio that increases with decreasing AC, as shown in Fig.3(plot (a)). Among three acids, titration using HCl shows (Fig.3(plot (c)) –ve error% in both AC and ABC differentiation with the existence of a similar equilibrium error problem as found in back titration method. The error% of the both AC and ABC might have also from ACm formation in the first step using phenolpthalein titration where ACm converted to CAm directly, instead of ABC formation, as seen in equation 18’. This ABC and ACm equilibrium (eqn. 19’) at high end ratios of AC and ABC might have influenced to receive this high error% in the both –ve and þve. In higher ABC, more of ACm might have formed and vice versa at lower ABC that makes an equilibrium variation depending upon the concentration of ABC, which leads high error% in their determination. Further works can be carried out using HCl acid.
331
0.2 0.0 -0.2
(NH4)2CO3 NH4HCO3
-0.4 -0.6 -0.8 2:8
4:6
5:5
6:4
7:3
8:2
9:1
Concn. by ratio Fig. 5. Effect of (NH4)2SO4 with different ratios of (NH4)2CO3/(NH4)HCO3 mixture on their identification by double indicator method. Added (NH4)2SO4 concentration¼ 0.15 M.
except high end ratio of 9:1, indicating effective replacement of the Na þ ion in ABC. The reason why the minimized error occurred is, only a small amount of AC becomes ACm that might not converted to SC while addition of SS due to sodium ammonium carbamate and double salt formation as shown in eq. 23. But one positive effect of the presence of sodium sulfate is that it enables the differentiation of ABC and AC without its interference. Nonetheless, more accurate differentiation is required regardless of ABC/AC ratio. Na2SO4 þNH2COONH4-NH2COONaþNH4.Na.SO4
(23)
3.4. Approach through the effect of equilibrium control by addition of salt and temperature As explained by eqs. 1 and 2, a common ion in the equilibrium change is sodium ion. One possibility of minimizing equilibrium change involves the addition of excess ammonium ion to the titration solution, here ammonium sulfate is probably a good choice, as shown by eq. 25. Ammonium sulfate added titration results to identify the ABC and AC are depicted in Fig. 5. There appears the error% for both þve and –ve errors has controlled within in 3 71% upto the ratios of 2:8 to 6:4 of AC and ABC. Also, there might be an equilibrium arrest between AC and ACm if add the AS (eqn. 24) for this both þve and –ve error% minimization. Beyond these ratios, the both -ve and þve error% were not controlled within the error limit. These results explain higher the AC concentration in the reaction mixture makes higher the both –ve and þve error%, which means the equilibrium only between AC and ABC at high concentration. In other words, at high ratios of AC, AS is not able to control the conversion to ABC, as shown in eq. 25. But, the both –ve and þve error% are almost controlled in all the ratios of AC and ABC when compared with no salt addition and in presence of Na2SO4, as shown in Fig.7. (NH4)2SO4 (24) NH2COONH4 (NH4)2CO3 At higher AC ratio:
-0.8 2:8
4:6
5:5
6:4
7:3
8:1
9:1
Concn. by ratio Fig. 4. Replacement titration with Na2SO4 on different ratios of (NH4)2CO3/(NH4) HCO3 mixture on their differentiation by double indicator method. Added Na2SO4 concentration¼0.14 M.
(NH4)2CO3(n) þH2Oþ NH4HCO3(n)-(NH4)2CO3(n-1) þ(NH4)HCO3 (n þ 1) þNH4OH(n)
(25)
It is seen from eq. 13 that there is a possibility of AC decomposition to ABC via H2O at high concentration ratios and forms NH4OH in
332
M. Govindan et al. / Talanta 132 (2015) 327–333
0.6
ACm
AS
AC
0.4
AS 5 °C
ABC
Found difference (g)
Scheme 1. Equilibrium arrest between AC, ACm, and ABC at different conditions.
0.2 0.0 -0.2 -0.4 -0.6 -0.8 25
20
15
10
5
ο
Temperature ( C) Fig. 6. Effect of temperature with high ratios of (NH4)2CO3/(NH4)HCO3 mixture on their differentiation in presence of (NH4)2SO4 by double indicator method. Added (NH4)2SO4 concentration¼ 0.15 M. Insert figure: Concentration difference effect in absence of (NH4)2SO4 at 5 1C on different ratios of (NH4)2CO3/(NH4)HCO3.
increasing AC to ABC ratios, the differentiation error% is increased up to 20%, which indicates AS is the only component responsible to arrest or control the equilibrium transfer between AC to ACm, and AC to ABC, as shown in Scheme 1. The both AS presence (AC and ACm equilibrium arrested) and low temperature (AC and ABC equilibrium arrested) give highly synergistic effect for clear differentiation of AC and ABC at all ratios. For clear differentiation, comparative results are plotted and presented in Fig. 7, here shown how the error trend of AC and ABC is minimized at different experimental conditions. It clearly explains the accurate identification ABC and AC is only possible by titration in the presence of AS at 5 1C, especially along with impurities. We believe that this present condition may support to get more accurate identification when analyze through an instrument. At 5 1C in presence of AS in all ratios: 3.5. Validation through real sample
Fig. 7. Overall comparitive plot on differentiation of ((NH4)2CO3/(NH4)HCO3 ratios with different experimental conditions using double indicator method: (a) without addition of salts; (b) With Na2SO4; (c) With (NH4)2SO4; (d) At 5 1C with (NH4)2SO4.
solution or NH3 might have liberated from solution. Moreover, the concentration we took is very less (0.1 M) for titration, which makes no loss of NH3 by liberation. We tried to check the pH value after phenolpthalein titration, but found no change or high pH value. It is well known that the reaction can be arrested at low temperature [34], and thus, there is a possibility that the conversion of AC could be controlled by decreasing temperature. As a first trial, the higher AC ratios were considered to check temperature effect and the obtained results are presented in Fig. 6. The five different temperatures examined from 25 to 5 1C shows a clear decrease in the both –ve and þve error%. Irrespective of the higher three ratios, the both –ve and þve error% decreased with decreasing temperature to within an error% of (3 71). Nonetheless, only low temperature (5 1C) with absence of AS shows ca. 20% error in both þve and –ve in all the ratios, as shown in the insert figure of Fig.6. At lower ratios. 2:8 and 4:6, the ABC differentiation is very effective (within 3% error) at 5 1C in the absence of AS and it starts increase with
The desulfurized crude sample was first purified by adding water and then reacted with NH3 and CO2 to produce NaHCO3. In order to finalize the reaction conditions, ABC and AC determination are important in a secondary waste solution of AS, as shown in reaction 26. The total reaction process was carried out as shown in Fig.1. In brief, 11.1 Kg of crude desulfurised waste were dissolved in 24.12 L of water. The slurry solution heated to 451C along with stirring about 30 min. Then the slurry was then transferred to filter press machine, which was operated at 451C and 4 atm pressure. Typically 11 wt% of insoluble impurities were collected from the filter press. The filtered solution has analyzed by titration for carbonate and bicarbonate, ISE for sodium ion, IC for both cations and anions. In these given conditions, approximately 90 wt% (obtained salt weight/(total raw waste weight-impurities weight) x 100) sodium sources through filter press method have been recovered. The exact weight and volume that obtained were depicted in row 7 in Table 1. The analyzed filtered desulfurized solution was then sent to carbonation tank to produce NaHCO3 using ammonia solution and CO2 gas. Based on the analysis value of sodium source, the ammonia solution 0.9 mol ratio against sodium source was taken for the production of sodium bicarbonate. Under optimized conditions of temperature, water volume, and pressure, sodium bicarbonate production process were done using the carbonation tank. Then the reaction slurry sent to peeler type decanter modified with the overhead temperature controller. The peeler type decanter was operated at 45 1C temperatures both on overhead and reactants slurry, 800 mesh filter and 3500 rpm. The product error was predicted based on the analyses of both solid and secondary waste liquid by titration, UV-Visible spectroscopy, and ISE methods. The total ions of both solid and liquid that found are displayed in row 11–12 in Table 1. The found difference of BC and C in the presence of all is 27 0.5%. This indicates validation of the present method and very crucial to finalize the further modification in perfection process. Crude wasteþNH3 þCO2 þH2O-NaHCO3 þ (NH4)2SO4
(26)
4. Conclusions The AC and ABC were identified within an error% successfully by simple titration. Back titration results confirm it will differentiate
M. Govindan et al. / Talanta 132 (2015) 327–333
333
Table 1 Analysis results from regeneration reaction process of NaHCO3 in desulfurized waste. Raw sample weight(Kg) Water volume (L) Separation temperature Impurity weight (Kg) and % Filtrate soln. obtained (L) Sodium sources wt (Kg) and wt%
For preparation of NaHCO3 Water volume for NaHCO3 production (L) Reaction tempr. (1C) and Time (min) Needed CO2 (Kg) ammonia (Kg) and vol. (L) Product weight (Kg) and wt% Analysis results Ions name SO24 CO23 HCO3 NH4þ Na þ
11.1 24.12 45 1C 1.21; 10.92% 23.12 Na2SO4 ¼ 7.66 Na2CO3 ¼ 0.99 NaHCO3 ¼0.33 Total weight¼ 8.98 Yield in %¼ 8.98/9.89(11.1-1.21¼ 9.89) ¼90.78 23.12 35–38 and 60 min NH3 (4.519 Kg) or NH4OH solution (8.319 L): Need CO2 (11.677 Kg) 8.34; 84.33(8.34/9.89) Reactants weight (Kg) Found weight in solid (Kg) Found weight in solution (Kg) 5.178 0.119 4.904 0.559 0.018 0.531 6.157 5.585 0.559 2.324 0.022 2.161 2.999 2.177 0.805
only one ratio of AC and ABC (6:4) and leave a high error to all other ratios makes unsuitable in their complete identification. Also, the double indicator titration method confirms, similar as the back titration method, not suitable to identify the AC and ABC except one ratio (5:5). In a replacement titration through the addition of Na2SO4 minimizes the identification of AC and ABC error% a less than the only double indicator method confirms partial equilibrium arrest between ACm and SC and conversion of AC and ABC to SC and SBC. Instead of SS, AS addition give 37 1 error% up to the ratios of 6:4 confirms the effective minimization in complete equilibrium arrest between ACm and AC, and conversion of AC to ABC. Beyond this 6:4 ratio, the error% has increased with increase in ABC concentration, which concludes only equilibrium between AC and ABC conversion. Finally, at low temperature 5 1C, the error% is completely arrested all equilibriums and minimized the error%, even at high ratios 7:3 of AC and ABC. The validation of the present method in real sample analysis is perfectly working to identify the BC and C in the presence of excess AS within an allowed error%. These results conclude that low temperature in the presence of excess AS is more suitable to differentiate the AC and ABC by titration analysis. Acknowledgements This study was supported by the Korea Research Foundation and the Korean Federation of Science and Technology Societies funded by Korea Government (MOEHRD, Basic Research Promotion Fund) and the National Research Foundation of Korea (NRF) funded by the Korean government (MEST) (Grant Nos. 20100027330 & 2014001974). Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.talanta.2014.08.044. References [1] [2] [3] [4]
I. Dahlan, K.T. Lee, A.R. Mohamed, J. Sci. Ind. Res 163 (2013) 83–87. R.-B. Lin, S.-M. Shih, C.-F. Liu, Chem. Eng. Sci. 58 (2003) 3659–3668. C.F. Liu, S.-M. Shih, Ind. Eng. Chem. Res. 45 (2006) 8765–8769. C. Wu, S.-J. Khang, T.C. Keener, S.-K. Lee, Adv. Environ. Res. 8 (2004) 655–666.
% error 3.08 1.85 2.28 5.24 1.83
[5] L. Xu, J. Guo, F. Jin, H. Zeng, Chemosphere 62 (2006) 823–826. [6] R. Sundararaman, C. Song, Energ. Fuel. 27 (2013) 6372–6376. [7] H. Deng, H. Yi, X. Tang, H. Liu, X. Zhou, Ind. Eng. Chem. Res. 52 (2013) 6778–6784. [8] I. Dahlan, K.T. Lee, A.H. Kamaruddin, A.R. Mohamed, Environ. Sci. Technol. 40 (2006) 6032–6037. [9] G. Muthuraman, S.J. Chung, I.S. Moon, J. Ind. Eng. Chem. 51 (2012) 2697–2703. [10] K.C. Pillai, S.J. Chung, T. Raju, I.S. Moon, Chemosphere 76 (2009) 657–664. [11] J. Bichel, S. Stephen, Method for recovering purified sodium bicarbonate and ammonium sulfate. U.S. Patent, 0156775 A1, Aug. 12, 2004. [12] F. Christensson, H.C.S. Koeford, A.C. Petersen, K. Rasmussen, Acta Chem. Scand. A 32 (1978) 15–17. [13] W.C. Femelius, in: W.C. Femelius (Ed.), Inorganic Syntheses, Vol. II, McGrawHill Book Company Inc, New York, 1946, pp. 85–86 (Chapter 4). [14] J.T. Yeh, K.P. Resnik, K. Rygle, H.W. Pennline, Fuel Process. Technol. 86 (2005) 1533–1546. [15] J.F.I. Thomas, J.J. Lynch, J. Amer. Water Works Assoc 52 (1960) 259. [16] A.A.B. Pichle, M. Cefola, Ind. Eng. Chem 6 (1939) 327–332. [17] A. Suchier, Angew. Chem. 44 (1931) 534–536. [18] R Belcher, A.J. Nutten, A.M.G. Macdonald, Quantitative Inorganic Analysis, Third ed., Butterworths, London, 1970. [19] C.L. Wilson, D.W. Wilson, Comprehensive Analytical Chemistry, Elsevier, London, 1960 (Classical Analysis). [20] N. Wen, M.H. Brooker, J. Phys. Chem. 99 (1995) 359–368. [21] L. Meng, S. Burris, H. Bui, W.P. Pan, Anal, Chem 77 (2007) 5947–5952. [22] Q. Zhao, S. Wang, F Qin, C. Chen, Ind. Eng. Chem. Res. 50 (2011) 5316–5325. [23] F. Mani, M. Peruzzini, P. Stoppioni, Green Chem. 8 (2006) 995–1000. [24] R. Bosco, B. Grandjean, Using raman spectroscppy to control carbonate/ bicarbonate concentrations, U.S. 2011, 0125326 A1, May 26, 2011. [25] P.L. Fosbøl, K. Thomsen, E.H. Stenby, Ind. Eng. Chem. Res. 48 (2009) 4565– 4578. [26] M. Laliberte, J. Chem. Eng. Data. 52 (2007) 321–335. [27] H. Huang, S.G. Chang, Energ. Fuel. 16 (2002) 904–910. [28] W.T. Bolleter, C.J. Bushman, P.W. Tidwell, Anal. Chem. 33 (1961) 592–594. [29] S.A. Ben Mussa, H.S. Elferjani, F.A. Haroun, F.F. Abdelnabi, Int. J. Pharm.Tech. Res 1 (2009) 598–604. [30] F. Qin, S. Wang, I. Kim, H.F. Svendsen, C. Chen, Int. J. Green. Gas Con 5 (2011) 405–412. [31] G.G. Wildgoose, A.T. Masheter, A. Crossley, J.H. Jones, R.G. Compton, Int. J. Electrochem. Sci. 2 (2007) 809–819. [32] M. Gronvald, C. Faurhold, Acta Chem. Scand. 14 (1960) 1374–1380. [33] P.D. Wimberley, O. Siggaard-Anderson, N. Fogh-Anderson, A.B.T. Boink, Scand. J. Lab. Inv. 45 (1985) 7–10. [34] G. Beck, J. Janek, Solid State Ionics 170 (2004) 129–133.
