Mercury Absorption in Aqueous Oxidants Catalyzed by Mercury(II)

Department of Chemical Engineering, The University of Texas at Austin, Austin, Texas 78712-1062 ... tion of Hg(II) and dichromate absorbed mercury mor...
0 downloads 0 Views 106KB Size
Download PDF
380

Ind. Eng. Chem. Res. 1998, 37, 380-387

Mercury Absorption in Aqueous Oxidants Catalyzed by Mercury(II) Lynn L. Zhao† and Gary T. Rochelle* Department of Chemical Engineering, The University of Texas at Austin, Austin, Texas 78712-1062

The absorption of elemental Hg vapor into aqueous solution containing Hg(II) was measured in a stirred cell contactor at 25 and 55 °C. In 0.8 M HNO3, the reaction is first-order in Hg and Hg(II), respectively. The overall second-order rate constant is given by k2 ) 2.90 × 109 × exp(-1765/T). In 0.8 M HNO3 with the addition of H2O2, the reaction is first-order in Hg, H2O2, and Hg(II), respectively. The overall third-order rate constant is given by k3 ) 2.13 × 1023 × exp(-10110/T). The addition of Fe2+ or Fe3+ has no immediate effect on mercury removal. In 0.8 M HNO3 with the addition of K2Cr2O7, the reaction is first-order in Hg, Cr2O7d, and Hg(II), respectively. The overall third-order rate constant is 4.3 × 108 M-2 s-1 at 25 °C. For mercury absorption in Hg(II) obtained by HgCl2 injection, the presence of HNO3 greatly enhanced Hg absorption. H2SO4 had a comparable positive effect while HCl had a negative effect. Succinic acid-NaOH buffer solution greatly enhanced Hg absorption in Hg(II), but NaHCO3-NaOH inhibited Hg absorption in Hg(II). MnSO4 mildly enhanced Hg absorption in Hg(II). At MnSO4 concentrations lower than 0.22 M, a constant overall third-order rate constant of 4.4 × 107 M-2 s-1 was obtained at 25 °C. NaCl, MgSO4, FeCl3, CaCl2, and MgCl2 all inhibited Hg absorption in Hg(II). Under most conditions, oxygen in the gas phase did not have any effect on Hg absorption in Hg(II). However, oxygen had a positive effect on Hg absorption in Hg(II) when HCl or NaHCO3/NaOH was present in the solution. Introduction EPA method 29 for field sampling of mercury specifies the use of hydrogen peroxide and nitric acid in the first impinger and permanganate and sulfuric acid in the following impinger (Environmental Protection Agency, 1992). This method assumes that hydrogen peroxide and nitric acid absorb divalent mercury but not elemental mercury vapor. However, results obtained at the High Sulfur Test Center indicate that the EPA mercury speciation method may not be reliable (Peterson et al., 1995). Several other researchers have indicated that the oxidation of elemental mercury by hydrogen peroxide may be important in the atmospheric environment (Lindqvist et al., 1984; Brosset, 1987; Kobayashi, 1987). More recently, a German company has commercialized a sulfur dioxide and mercury scrubbing process called MercOx (Parkinson, 1996). In their process, a 35% aqueous solution of hydrogen peroxide is sprayed into the flue gas. Elemental mercury is oxidized by hydrogen peroxide to Hg(II) and remains in the solution. In a parallel reaction, the water spray converts SO2 to sulfuric acid. As a result, all gaseous pollutants are converted and trapped in the solution. Dissolved mercury is removed by ion exchange, and the acids are neutralized to salts and precipitated gypsum. In a recent pilot test of treating 500 m3/h of flue gas stream at a sludge incinerator, mercury concentrations were reduced from a few hundred to 20 µg/m3, well below the 50 µg/m3 German limit. The combination of hydrogen peroxide and a polyvalent metal ion has also been suggested as a potential oxidizer for elemental mercury. Kobayashi (1987) stud* To whom correspondence should be addressed. E-mail: [email protected]. Telephone: 512-471-7230. Fax: 512-4757824. † Present address: R & D Division, BCIT, Exxon Chemical Company, Baytown, TX 77522.

ied the mixtures of hydrogen peroxide with thirteen metal ions (Fe3+, Mn2+, Zn2+, Pb2+, Cu2+, Co2+, Ni2+, Cd2+, Sb3+, Bi3+, VO3-, Ca2+, and Al3+). Among these, only Fe3+ enhanced mercury absorption and the reaction was first-order in hydrogen peroxide and Fe3+, respectively. Munthe and McElroy (1992), however, concluded that the reactions between aqueous elemental mercury and hydrogen peroxide, or Fe3+, or a mixture of the two, were not significant. Other researchers have studied the effect of Hg(II) on elemental mercury removal. Qualitative results obtained by Morita et al. (1983) reported that the rate of mercury absorption increased with increasing Hg(II) concentration and with increasing oxidation potential of dichromate solutions in sulfuric acid. The combination of Hg(II) and dichromate absorbed mercury more efficiently than either component alone. Strong mercury absorption in Hg(II)-H2SO4 was also observed (Morita et al., 1983). Allgulin (1974) found that a solution containing mercury(II) ions ranging from 0.02 g/L to saturation and at least one anion selected from the group Cl-, Br-, I-, and SO42- was effective for removing elemental mercury. The objective of this study was to measure the kinetics of elemental mercury reaction with hydrogen peroxide under acidic conditions in a well-characterized gasliquid contactor. We also investigated the role of mercury(II) ion, oxygen, and other oxidants and reagents in elemental mercury absorption in a variety of aqueous solutions. Experimental Methods The detailed description of the stirred tank reactor can be found in a previous paper (Zhao and Rochelle, 1996). Gas containing Hg was contacted with aqueous solutions over an agitated, unbroken interface of 81 cm2.

S0888-5885(97)00155-3 CCC: $15.00 © 1998 American Chemical Society Published on Web 01/16/1998

Ind. Eng. Chem. Res., Vol. 37, No. 2, 1998 381

manganate (Zhao and Rochelle, 1996) and its value is listed in Table 1. In all of the experimental results, the Hg partial pressure at the gas-liquid interface, PHgi, is calculated from the experimental flux by

PHgi ) PHgb -

NHg kg,Hg

(1)

For most experiments the results are reported as normalized flux, Kg′, analogous to the overall gas-phase mass-transfer coefficient: Figure 1. The stirred tank reactor system for mercury absorption. Mercury permeation tube is inside the U-tube immersed in a water bath. TC is temperature controller, FM is mass flowmeter.

Kg′ )

NHg PHgi

(2)

Table 1. Physical Properties of Hg and Hg2+ a

(cm2/s)

DHg-H2O DHg2+-H2Ob (cm2/s) HHgc (atm/M) kg,Hg (mol/(s atm m)) typical value k°l,Hg (m/s) typical value k°l,Hg2+ (m/s) typical value

25 °C

55 °C

10-5

10-5

1.19 × 8.47 × 10-6 8.91 0.0344(ng)0.38 0.4 2.42 × 10-7(nl)0.73 2.6 × 10-5 2.04 × 10-7(nl)0.73 2.2 × 10-5

2.21 × 1.65 × 10-5 35.64 0.0344(ng)0.38 0.4 7.64 × 10-7(nl)0.64 4.7 × 10-5 6.59 × 10-7(nl)0.64 4.0 × 10-5

a Estimated by method of Sitaraman et al. (1963). b Obtained from Lide (1994). c Obtained from Clever et al. (1985).

Figure 1 gives the Teflon-coated stirred tank system used for mercury absorption. Mercury was obtained from a permeation tube with a nitrogen flow of 100 cm3/ min. In a typical experiment, mercury diluted with nitrogen (1 L/min) bypassed the reactor. The gas flows were maintained by mass flow controllers. After the mercury analyzer gave a reasonably stable reading (510 min), the Hg-N2 stream was shifted to flow above 0.8 M nitric acid or another background solution, such as other acids, buffer solution, or water. Again after the analyzer reached a stable reading, a known amount of 30 wt % hydrogen peroxide solution or mercuric chloride solution prepared from solid was injected using a syringe with a long needle into the reactor and the outlet mercury concentration was recorded continuously. Each active liquid-phase ingredient was injected sequentially rather than mixed outside the reactor. Prior to each injection, the same amount of solution was extracted from the reactor to keep the final liquid volume constant during mercury absorption. The reactor outlet gas was immediately diluted with nitrogen by a factor of 10 at 25 °C and 15 at 55 °C to minimize water vapor interference on mercury analysis. A three-point calibration of the analyzer (cold vapor atomic absorption) was conducted both before and after the experiment. The mercury concentration obtained when the Hg-N2 was passed over the background solution of the actual experiment was used as the inlet mercury concentration (wet calibration). The rate of mercury absorption was calculated from the gas-phase material balance. The bulk hydrogen peroxide concentration was determined by titration with potassium iodide and certified 1 N sodium thiosulfate solution. HgCl2 or other reagent additions to the liquid phase were determined by weighing the syringe before and after the injection. The specifications of the reagents are given in Zhao (1997). The gas-phase mass-transfer coefficient, kg,Hg, was obtained from mercury absorption in concentrated per-

The value of k°l,Hg (Zhao, 1997) obtained from mercury desorption experiments in an identical reactor was modified to give k°l,Hg2+:

x

k°l,Hg2+ ) k°l,Hg

DHg2+-H2O

(3)

DHg-H2O

Similarily, the Hg2+ concentration at the gas-liquid interface [Hg2+]i is calculated from the experimental flux by

either [Hg2+]i ) [Hg2+]b,initial injected + [Hg2+]b,absorbed +

NHg (4) k°l,Hg2+

or [Hg2+]i ) [Hg2+]b,initial injected + [Hg2+]b,absorbed -

NHg (5) k°l,Hg2+

depending whether Hg2+ is accumulated or consumed at the gas-liquid interface. The portion of [Hg2+]b resulting from mercury absorption was obtained by the integration of the gas-phase material balance. The nominal concentration of injected mercuric chloride was used for the portion of [Hg2+]b resulting from mercuric chloride injections. All other physical constants used in the calculation are tabulated in Table 1. Results and Discussion Absorption in Mercuric Chloride. With only distilled water in the liquid phase, injections of HgCl2 resulted in step decreases of the outlet mercury concentration, as shown in Table 2. As a result, the normalized flux, Kg′ increases with HgCl2. Experimental results in Table 2 indicate that the injection of more HgCl2 resulted in lower solution pH, as pH was reduced from 4.82 with 0.008 mM HgCl2 to 4.14 with 1 mM HgCl2. Lower pH enhanced mercury removal by Hg(II), as shown in later discussions. Effects of SaltssNaCl, MgSO4, FeCl3, CaCl2, and MgCl2. The effects of five different salts on mercury absorption in Hg(II) were investigated by sequentially injecting HgCl2 and salts. Figure 2 gives the third-order rate constant k3′ vs [Hg(II)]i. k3′ was obtained from

NHg )

PHgi HHg

xk ′D 3

2

Hg-H2O[Hg(II)]i

(6)

382 Ind. Eng. Chem. Res., Vol. 37, No. 2, 1998 Table 2. Hg Absorption in Water with Sequential HgCl2 Injection at 25 °Ca O2 (%)

PHgi × 108 (atm)

NHg × 109 (mol/(s m2))

Hg2+ injected (mM)

pH

0 0 0 0 0 0 0 0 0 0 20 20 20 20

10.1 9.7 7.6 5.8 4.0 3.1 2.1 10.0 8.0 1.6 9.2 7.9 5.8 2.7

0.06 0.40 1.84 3.10 4.39 4.97 5.67 0.16 1.59 6.08 0.71 1.63 3.13 5.26

0.006 0.022 0.091 0.195 0.395 0.593 1.164 0.008 0.068 1.157 0.030 0.080 0.212 0.694

4.82 4.45 4.14 5.10 4.90 4.74 4.51

a

Total Hg-N2 or Hg-air flow rate was 1 L/min. PHg,in was 1.02 × 10-7 atm. kg,Hg was 0.39 gmol/(s atm m2). k°l,Hg2+ was 2.2 × 10-5 m/s at 25 °C.

Figure 3. Hg absorption in 0.8 M HNO3 with sequential HgCl2 injection at 25 and 55 °C. Total Hg-N2 flow rate was 1 L/min.

between elemental mercury vapor and mercury(II) is given by the following mechanism: H+

Hg + Hg(II) 98 products

(7)

At constant pH the reaction rate is given with the second-order rate constant, k2:

reaction rate ) k2[Hg][Hg(II)]

(8)

Using approximate surface renewal theory (Danckwerts, 1970), the flux of elemental mercury, NHg, should be given by

NHg ) Figure 2. The effect of salts (concentrations in M) on Hg absorption at 25 °C. k3′ was obtained from eq 6. Filled squares represent data with 20% O2 while empty squares with N2 only. All other legends represent data with N2. Table 3. Effect of Oxygen and NaCl on Hg Absorption in NaCl at 25 °C NaCl (M) 1 1 1 0.1 0.1 0.1

O2 (%)

HgCl2 injected (M)

pH

results or Kg′ (mol/(s m2 atm))

20 20 20

8.8 × 10-6 5.5 × 10-5 3.5 × 10-4

6.78 6.77 6.72

0 0 Hgout < Hgina

0 0 0

5.2 × 10-6 1.3 × 10-4 3.5 × 10-4

5.8 × 10-4 1.9 × 10-2 4.6 × 10-2

a The HgCl injection caused the outlet Hg concentration to 2 decrease immediately but gradually increased to near the inlet Hg concentration. So the Kg′ value was very small in this case.

All of the above five salts inhibited the Hg removal ability of Hg(II). Oxygen up to 20% did not affect Hg absorption in water with HgCl2 injection (in the absence of salt). Table 3 gives the effect of oxygen and NaCl on Hg absorption in Hg(II) at 25 °C. The results indicate once again that NaCl inhibits Hg absorption in Hg(II) and the magnitude of the negative effect on Hg removal is proportional to the concentration of NaCl. Absorption by Mercury(II) with the Addition of a Strong Acid or Buffer Solutions. With a solution of a strong acid, we have found that the rate of reaction

PHgi kD [Hg(II)]i HHgx 2 Hg-H2O

(9)

Mercury Absorption in 0.8 M Nitric Acid. Elemental mercury vapor was absorbed into 0.8 M nitric acid with sequential injection of mercuric chloride. Without any other oxidants, injection of mercuric chloride into 0.8 M nitric acid resulted in a step decrease in outlet mercury concentration. The results are presented in Figure 3 and Table 4. The reaction is first-order in Hg and Hg(II), respectively. The kinetic data at 55 °C was difficult to obtain and unstable compared to that at 25 °C. This was probably due to the difficulties encountered in cleaning the reactor from residual HgCl2 using the reducing agent NaBH4. Sometimes, the reactor had to be first soaked in NaBH4 to get rid of the residual HgCl2 and then soaked in H2O2 to neutralize the effect of NaBH4. In contrast, soaking the reactor in NaBH4 before each experiment when conducting experiments at 25 °C did not cause any difficulties and abnormal results. As shown in Figure 3, when the injected HgCl2 exceeded 3 × 10-5 M at 55 °C, the normalized Hg flux tended to flatten out. With HgCl2 less than 3 × 10-5 M, the overall second-order rate constant k2 is (7.8 ( 2.1) × 106 M-1 s-1 at 25 °C and (13.4 ( 8.0) × 106 M-1 s-1 at 55 °C. The experimental data for k2 are represented by the following expression:

(-1765 T )

k2 ) 2.90 × 109 exp

(10)

At 10-5 M HgCl2, the normalized mercury flux, Kg′, was 3.5 × 10-2 with 0.8 M HNO3 and 1.6 × 10-3 (mol/(s

Ind. Eng. Chem. Res., Vol. 37, No. 2, 1998 383 Table 4. Effect of Hg(II) on Mercury Absorption in 0.8 M HNO3a PHg,i × 108 (atm)

NHg × 109 (mol/(s m2))

[Hg(II)]i (µM)

k2 × 10-6 (1/(M s))

25

9.3 8.2 7.3 8.0 6.3 5.0 4.2 7.9 6.1 4.9 3.9

0.6 1.4 2.0 1.6 2.8 3.7 4.2 1.7 2.9 3.7 4.4

0.65 2.77 6.64 3.75 17.07 41.08 89.77 4.09 17.41 46.68 110.76

4.6b 7.3 7.7 7.2 7.8 8.6 7.6 7.3 8.5 8.3 7.9

55

9.4 8.7 7.4 6.7 6.7 8.0 7.5 8.6 7.7 7.3 6.5

0.5 1.0 1.8 2.3 2.3 1.4 1.8 1.1 1.6 1.9 2.4

1.48 5.64 29.93 83.53 161.50 10.02 24.01 7.10 22.59 49.52 127.06

13.1 13.5 11.4 7.7b 4.0b 18.6 13.8 12.6 11.6 7.5b 6.0b

T (°C)

-7 atm. k 2 aP Hg,in was 1.0 × 10 g,Hg was 0.4 gmol/(s atm m ). k°l,Hg2+ was 2.2 × 10-5 m/s at 25 °C and 4.0 × 10-5 m/s at 55 °C. b Represents data excluded from rate constant regression.

Figure 4. The dependence of second-order rate constant k2 on [Hg(II)]i with various acids and buffers in the solution at 25 °C. k2 was obtained from eq 9. N2 was in the gas phase unless otherwise indicated. Experiments with O2 were with 20% (by volume) O2.

m2 atm)) without acid. This indicates that the presence of HNO3 greatly enhanced Hg absorption by mercury(II). Effects of HNO3, H2SO4, and HCl. Figure 4 gives the results at 25 °C when nitric acid or sulfuric acid was present in the solution. The baseline is absorption in water without acid addition. k2 was significantly lower with 0.01 M HNO3 and no acid than with 0.8 M HNO3. The difference was most significant at low Hg(II) and diminished at higher Hg(II). For example, at 10-5 M Hg(II), Kg′ of mercury increased by a factor of 32 with 0.8 M HNO3 compared to that with HgCl2 alone. While at 10-3 M Hg(II), Kg′ of mercury only increased by a factor of 1.5. The effect of HNO3 is apparent in a comparison of mercury absorption in 0.01, 0.1, and 0.8 M HNO3 with

sequential HgCl2 injection. k2 was lower in 0.01 M HNO3 than in 0.1 M HNO3 at the same Hg(II) level. Figure 4 also shows that k2 in 0.01 M HNO3 was greater than with no HNO3 at the same level of Hg(II). However, there was very little increase in k2 when HNO3 was increased from 0.1 to 0.8 M. At high HgCl2, concentration differences in HNO3 did not make much difference in k2. The results of HNO3 injections from 8 × 10-5 to 0.7 M indicate that at Hg(II) greater than 1 mM, increased Hg(II) was the primary reason for high mercury absorption compared to the increased HNO3 acidity. For the other acid systems studied, keeping a high and constant level of H+ was essential to obtain an overall second-order reaction between elemental mercury and Hg(II). When a high level of HNO3 existed, injections of HgCl2 did not change the solution pH significantly. However, when a small amount of HNO3 was present, injections of HgCl2 decreased the solution pH significantly. k2 increased as the injected amount of Hg(II) was increased. This was the effect of both increased HgCl2 and lower pH associated with increased HgCl2. This is the primary reason that k2 depends on Hg(II), as shown in Figure 4. The effect of 0.8 M H2SO4 on Hg absorption at 25 °C is comparable to that of 0.8 M HNO3. The effect of HCl was studied in a similar manner with only elemental mercury vapor and nitrogen in the gas. HgCl2 was sequentially injected into 0.8 M HCl at 25 °C to give 10-6-10-3 M HgCl2. The outlet Hg concentration exceeded the inlet Hg concentration and increased as the concentration of injected HgCl2 increased. The same phenomenon was observed when 0.9 M HCl was used in another effort. This indicates that the presence of HCl in HgCl2 not only did not remove mercury but reduced some of the injected HgCl2 to elemental form. HgCl2 was sequentially injected into 0.9 M HCl with 20% oxygen in the gas phase. Although a finite amount of mercury removal was detected, the presence of HCl inhibited Hg absorption in HgCl2, as shown in Figure 4. However, in the absence of oxygen, the outlet Hg concentration exceeded the inlet Hg concentration and increased as the amount of injected HgCl2 increased. Effect of NaOH-Succinic Acid Buffer. From the results with various degrees of HNO3 and HgCl2 in the solution, it seems that pH plays an important role. Thus a 0.01 M NaOH-0.01 M succinic acid buffer solution was used to keep a reasonably constant pH at 5.6 while subsequently injecting HgCl2. The results are shown in Figure 4. k2 was comparable to that with 0.8 M HNO3, even though the buffer solution pH was much higher. This indicates that 0.01 M NaOH-0.01 M succinic acid buffer greatly enhances Hg absorption in Hg(II). With 20% oxygen in the gas phase, k2 did not change significantly. This indicates that oxygen has negligible effect on Hg absorption in 0.01 M NaOH0.01 M succinic acid buffer with sequential HgCl2 injection. Effect of NaHCO3-NaOH Buffer. With only nitrogen and Hg in the gas phase, NaHCO3-NaOH buffer was used to maintain the solution pH at 9.49-9.56. HgCl2 injections into NaHCO3-NaOH buffer gave 10-6-10-3 M Hg(II). The outlet Hg concentration increased as the amount of HgCl2 injected increased, as shown in Table 5. For all the HgCl2 injections performed, the outlet Hg concentration exceeded the inlet Hg concentration. This

384 Ind. Eng. Chem. Res., Vol. 37, No. 2, 1998 Table 5. Effect of Oxygen on Hg Absorption in NaHCO3-NaOH Buffer Solution at 25 °C O2

HgCl2 (M) 10-6

0 0 0

10-4 10-3

20 20 20

2.17 × 10-5 1.47 × 10-4 5.26 × 10-4

Table 6. Effect of H2O2 and Hg(II) on Mercury Absorption in 0.8 M HNO3 at 25 °Ca

pH

results

9.56 9.49

Hgout > Hgin Hgout > Hgin Hgout > Hgin

9.63 9.61 9.55

Hgout ) Hgin Hgout ) Hgin Hgout > Hgin

indicates that the injected HgCl2 was somehow reduced to elemental mercury and got stripped out of the solution. The presence of 20% O2 contributed to inhibiting the reduction of injected HgCl2, but not enough to cause Hg absorption. When the injected HgCl2 was increased to a certain high value, such as 5.26 × 10-4 M, 20% O2 was no longer enough to prevent HgCl2 from being reduced. Mercury Absorption in Hydrogen Peroxide and Nitric Acid. The rate of reaction between elemental mercury vapor and hydrogen peroxide is given by the mechanism: H+

Hg + H2O2 98 Hg(II) + other products

PHg,in PHg,i × 108 [H2O2]b [Hg(II)]i × 107 NHg × 1010 k3 × 10-8 (ppb) (atm) (M) (M) (mol/(s m2)) (1/(M2 s)) 20.1

2.1 2.1 2.1 2.1 2.1 2.0 2.0 2.0

0.1 0.1 0.1 0.1 0.1 0.4 0.4 0.4

97.5

7.8 7.8 7.3 6.7 6.6 9.7 8.4 8.2 8.0 7.3 7.0 6.8 9.9 9.9 9.1 8.8 8.3 7.9 5.7 3.3 3.3 1.5

1.0 1.0 1.0 1.0 1.0 0.2 0.6 0.6 0.6 0.9 0.9 0.9 0.1 0.1 0.3 0.3 0.6 0.6 0.3 0.3 0.3 0.3

(11)

Since the reaction product Hg(II) catalyzes Hg absorption, the rate of reaction at constant pH with the third-order rate constant, k3, is given by

reaction rate ) k3[Hg][Hg(II)][H2O2]

(12)

Using surface renewal theory, the flux of elemental mercury, NHg, should be given by

NHg )

PHgi kD [Hg(II)]i[H2O2]i HHg x 3 Hg-H2O

(13)

k3 ) 2.13 × 10

23

-10110 exp T

(

)

(15)

At our conditions 0.1 mM Hg2+ was sufficient to get 50% gas film resistance. Therefore, there may be specific conditions where elemental Hg will be absorbed by the first impinger in EPA method 29. As a result

3.9 4.1 3.8 4.1 4.4 4.0 3.9 4.8

17.0 17.3 20.8 24.8 25.3 3.7 12.9 14.1 15.5 20.5 22.8 23.9 2.0 2.3 8.1 9.8 13.6 16.0 32.0 48.6 48.9 61.5

3.2 3.1 3.8 4.7 4.8 3.2 3.9 3.7 4.0 3.9 4.0 4.1 3.1 3.5 3.7 4.3 3.9 4.7 3.4 3.1 3.2 4.0

Table 7. Effect of H2O2 and Hg(II) on Mercury Absorption in 0.8 M HNO3 at 55 °Ca PHg,in PHg,i × [H2O2]b [Hg(II)]i NHg × 1010 k°l,Hg2+ × k3 × 10-8 (ppb) 108 (atm) (M) × 107 (M) (mol/(s m2)) 105 (m/s) (1/(M2 s)) 97.5

9.9 9.8 8.9 8.8 9.3 8.9 7.9 7.8 7.7 6.7 9.4 9.4 9.4

0.1 0.1 0.2 0.2 0.2 0.2 0.4 0.4 0.4 0.7 0.1 0.1 0.1

0.1 0.1 0.3 0.4 0.2 0.4 0.7 0.7 0.8 1.1 0.1 0.2 0.2

2.0 2.5 8.4 9.3 6.1 8.6 15.2 15.7 16.6 22.9 5.1 5.1 5.1

4.1 4.1 4.1 4.1 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0

90.6 90.4 86.6 97.3 86.0 82.3 84.4 81.8 89.5 92.6 86.2 86.2 86.7

97.6

9.7 9.6 9.5 8.6 8.3 8.0 7.5 7.4 6.8 6.4 5.7

0.1 0.1 0.1 0.3 0.3 0.3 0.4 0.4 0.6 0.6 0.8

0.1 0.1 0.2 0.4 0.6 0.8 0.9 1.0 1.3 1.5 2.2

3.3 4.2 4.7 10.4 12.8 14.6 17.5 18.4 22.3 24.9 29.3

4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0

82.4 86.7 85.2 81.4 88.8 91.1 85.9 87.1 86.8 91.2 87.6

(14)

Hydrogen Peroxide and 0.8 M Nitric Acid. In a typical experiment, H2O2 was injected into 0.8 M HNO3. The Hg outlet concentration dropped when additional H2O2 was injected. It kept decreasing as absorption proceeded. This phenomenon can be explained by the accumulation of absorbed mercury species (mainly Hg(II)). This assumption was verified by the results with HgCl2 injection. The Hg outlet concentration dropped dramatically when HgCl2 was injected. This indicates that Hg(II) catalyzes Hg absorption in H2O2-HNO3. The experimental data are tabulated in Tables 6 and 7 for experiments conducted at 25 and 55 °C, respectively. Figure 5 plots the normalized flux of mercury, Kg′, versus the product of [H2O2]i and [Hg2+]i. The thirdorder rate constant is (4.0 ( 1.1) × 108 M-2 s-1 at 25 °C and (88.8 ( 9.9) × 108 M-2 s-1 at 55 °C. The experimental data for k3 are represented by the following expression:

1.1 1.1 1.5 1.9 2.0 0.2 0.7 0.9 1.1 1.6 1.9 2.1 0.1 0.1 0.5 0.6 0.8 1.0 24.6 183 183 1090

0.1 0.2 0.2 0.2 0.3 0.6 0.7 0.8

a Hg-N flow ) 1 L/min. k 2 2 g,Hg ) 0.4 gmol/(s atm m ). k°l,Hg2+ ) 2.2 × 10-5 m/s. [H2O2]b was determined by titration.

where

[H2O2]i ≈ [H2O2]b

0.01 0.01 0.02 0.02 0.02 0.04 0.1 0.1

a Total Hg-N flow rate was 1 L/min. k 2 g,Hg was 0.4 gmol/(s atm m2). [H2O2]b was determined by titration.

mercury speciation by sequential absorption in peroxide and permanganate impingers may not give accurate or even reproducible results. In particular, the anomalous results on Hg injection at Shawnee (Peterson et al., 1995) may be explained as absorption of elemental mercury in the first H2O2-HNO3 impinger. Effects of FeCl3 or FeSO4 in H2O2 and HNO3. The effects of Fe3+ and Fe2+ were investigated by adding

Ind. Eng. Chem. Res., Vol. 37, No. 2, 1998 385

Figure 5. The dependence of Kg′ of mercury on the product of [H2O2]i and [Hg(II)]i during mercury absorption in H2O2-0.8 M HNO3. The inlet Hg was tested at 20 and 9 ppb. The injected H2O2 was varied between 0.1 and 1.0 M while Hg2+ was varied between 0 and 1.1 × 10-4 M.

FeCl3 or FeSO4 to H2O2 with and without 0.8 M HNO3. External HgCl2 injection into the above-described systems was tested in some of the experiments, both before and after FeCl3 or FeSO4 was injected. From 4.9 × 10-4 to 0.4 M FeCl3 was injected into 0.1-0.4 M H2O2 with or without 0.8 M HNO3. No significant Hg absorption was observed due to the injection of FeCl3. In a similar manner, from 2.4 × 10-5 to 0.2 M FeSO4 was injected into 0.1-0.5 M H2O2 with or without 0.8 M HNO3. No significant Hg absorption was observed due to the injection of FeSO4. In all the experiments, HgCl2 injection, no matter when it was injected, always caused immediate drop in the outlet Hg concentration. Thus it was concluded that Fe3+ or Fe2+ had no immediate effect on elemental mercury removal. In contrast, as little as 10-6 M HgCl2 resulted in dramatic elemental mercury absorption. Mercury Absorption in Mercury(II) and Potassium Dichromate. The rate of reaction between elemental mercury vapor, Hg(II), and potassium dichromate in 0.8 M nitric acid is given by the following mechanism: H+

Hg + Hg(II) + Cr2O72- 98 products

(16)

When a fixed amount of HNO3 and Cr2O72- are present in the solution, the reaction rate can be written as

reaction rate ) k2[Hg][Hg(II)]

(17)

where k2 is the pseudo-second-order rate constant. Using surface renewal theory (Danckwerts, 1970), the flux of elemental mercury, NHg, should be given by

PHgi NHg ) kD [Hg(II)]i HHg x 2 Hg-H2O

(18)

Table 8. Mercury Absorption in K2Cr2O7-0.8 M HNO3 with Sequential HgCl2 Injection at 25 °Ca PHg,i × 108 (atm)

[Hg(II)]i × 105 (M)

NHg × 109 (mol/(s m2))

k2 × 10-7 (1/(M s))

7.1 6.3 5.5 3.9 2.5 1.9

0.4 0.9 1.8 6.1 14.7 33.0

2.2 2.7 3.3 4.4 5.4 5.8

1.5 1.4 1.3 1.4 2.1 1.9

a Total Hg-N flow rate was 1 L/min. The inlet mercury was 2 97.6 ppb. Solution pH ranged from 0.37 to 0.47. kg,Hg was 0.4 gmol/ (s atm m2). k°l,Hg2+ was 2.2 × 10-5 m/s. k2 is the pseudo-secondorder rate constant with a constant K2Cr2O7 concentration of 32.3 mM.

either component alone, as shown in Figure 6. The presence of HNO3 further enhanced Hg absorption. In addition, 0.8 M HNO3 has more positive effects on Hg removal than increasing K2Cr2O7 concentration alone. Solutions with all three componentssHg(II), HNO3, and K2Cr2O7swere the most efficient for mercury absorption compared to those with either one or two ingredients only. Of different combinations of Hg(II), K2Cr2O7, and HNO3 tested, only K2Cr2O7-0.8 M HNO3 with sequential HgCl2 injection exhibited a simple pseudo-secondorder reaction behavior. The data are tabulated in Table 8. The pseudo-second-order rate constant was 1.4 × 107 M-1 s-1 at 25 °C. Mercury Absorption in Mercury(II) and Manganese(II) Sulfate. The rate of reaction between elemental mercury vapor, Hg(II) and manganese(II) sulfate is given by the following mechanism:

Hg + Hg(II) + Mn(II) f products

(19)

Experimental results show that K2Cr2O7 alone (without Hg(II) or HNO3) removes little (if any) Hg, even with K2Cr2O7 as high as 39 mM. However, the combination of K2Cr2O7 and HgCl2 absorbs Hg more efficiently than

(20)

The rate of reaction is:

reaction rate ) k3[Hg][Hg(II)][Mn(II)]

In the above equations,

[Cr2O72-]i ≈ [Cr2O72-]b

Figure 6. The effect of K2Cr2O7 and HNO3 on Hg absorption at 25 °C. k2 was obtained from eq 18. Legends indicate the initial concentrations of K2Cr2O7 (mM) and HNO3 (M), respectively.

(21)

where k3 is the third-order rate constant. Using surface renewal theory (Danckwerts, 1970), the flux of elemental mercury, NHg, should be given by

NHg )

PHgi kD [Hg(II)]i[Mn(II)]i HHg x 3 Hg-H2O

(22)

386 Ind. Eng. Chem. Res., Vol. 37, No. 2, 1998 Table 9. Effect of Hg(II) and Mn(II) on Mercury Absorption in HgCl2-MnSO4 at 25 °Ca PHg,i × 108 NHg × 109 [Hg(II)]i × 105 [Mn(II)]b k3 × 10-7 (atm) (mol/(s m2)) (M) (M) pH (1/(M2 s))

Figure 7. absorption in water with sequential HgCl2 and MnSO4 injection at 25 °C. The injected Hg(II) ranged from 17 to 42 µM while Mn(II) ranged from 40 to 380 mM.

where

[Mn(II)]i ≈ [Mn(II)]b

(23)

Figure 7 gives the results of mercury absorption in distilled water with sequential HgCl2 and MnSO4 injections at 25 °C. No nitric acid or other reagent was present in the solution. The additions of MnSO4 enhanced mercury absorption in Hg(II). However, the enhancement was not significant since substantial amounts of MnSO4 were needed. Quantitative results are listed in Table 9. At MnSO4 less than 0.22 M, a third-order rate constant of 4.4 × 107 M-2 s-1 was obtained. A rate constant of 3.3-3.9 × 107 M-2 s-1 was obtained for MnSO4 greater than 0.23 M.

Table 10 gives a summary of rate constants obtained for the various aqueous systems. The second- and thirdorder rate constants, k2 and k3, are calculated from

PHgi kD [Hg(II)]i HHgx 2 Hg-H2O

1.7 1.9 2.2 2.7 3.5 3.7 3.8 3.9 2.7 3.0 3.1 3.3 3.7 3.8 3.9 3.9

1.8 1.7 1.7 3.5 3.9 3.8 3.7 3.6 2.9 3.0 2.9 2.9 4.2 4.2 4.1 4.0

0.04 0.06 0.08 0.08 0.17 0.22 0.28 0.38 0.10 0.13 0.15 0.19 0.23 0.26 0.31 0.35

3.89 3.81 3.72 3.61 4.04 3.96 3.90 3.84 3.76 3.72 3.66 3.61

4.4 4.2 4.1 4.3 4.7 4.4 3.9b 3.3b 4.3 4.3 4.4 4.4 3.9b 3.8b 3.6b 3.4b

a Total Hg-N flow rate was 1 L/min. The inlet Hg was 97.7 2 ppb. kg,Hg was 0.4 gmol/(s atm m2). k°l,Hg2+ was 2.2 × 10-5 m/s. b Represents data excluded from rate constant regression

Table 10. Summary of the Second- and Third-Order Rate Constants, k2 and k3, for Mercury Absorption in Different Aqueous Systems aqueous solution HgCl2-0.8 M HNO3 HgCl2-H2O2-0.8 M HNO3 HgCl2-K2Cr2O7-0.8 M HNO3 HgCl2-MnSO4

T °C

k2 × 10-6 (1/(M s))

25 55 25 55 25 25

7.8 ( 2.1 13.4 ( 8.0 14 (pseudo)

k3 × 10-8 (1/(M2 s))

4.0 ( 1.1 88.8 ( 9.9 4.3a 0.44

a Was measured at 32 mM K Cr O and various HgCl in 0.8 2 2 7 2 M HNO3.

Table 11. Comparison of Kg′ and k2 at 10-4 M Hg(II) and 25 °Ca

Summary of Results

NHg )

7.9 7.5 7.1 6.3 5.2 5.0 4.8 4.7 6.3 6.0 5.7 5.5 4.9 4.8 4.7 4.6

(24)

or

NHg ) PHgi kD [Hg(II)]i[other active reagent]i (25) HHgx 3 Hg-H2O Although the aqueous systems studied were diversified, most of them exhibited first-order dependence on Hg(II). To compare different systems, the Hg(II) concentration must be specified along with other active liquid or gas components. The effectiveness of the system for Hg removal is expressed as normalized flux, Kg′. The results are listed in Table 11. At 10-4 M HgCl2, 0.26 M H2O2-0.8 M HNO3 gives the best mercury removal, followed by 0.03 M K2Cr2O7-0.8 M HNO3, and then 0.1 or 0.8 M HNO3, or 0.8 M H2SO4. Conclusions Hg(II) catalyzed Hg0 absorption in aqueous solutions. Hg(II) alone in water removes Hg0 effectively with a reaction that is about second-order in Hg2+. The addi-

liquid and gas composition water, N2, or 20% O2 0.26 M H2O2-0.8 M HNO3 32 mM K2Cr2O7-0.8 M HNO3 3 mM K2Cr2O7 13 mM K2Cr2O7 39 mM K2Cr2O7 0.01 M HNO3 0.1 M HNO3 0.8 M HNO3 0.8 M H2SO4 0.01 M NaOH-0.01 M succinic acid 0.1 M MnSO4 0.011 M FeCl3 0.036 M MgCl2 0.1 M NaCl 0.02 M MgSO4 0.084 M CaCl2 0.9 M HCl, air 0.8 or 0.9 M HCl, N2 NaHCO3-NaOH buffer

Kg′ × 102 k2 × 10-6 b (mol/(s atm m2)) (1/(M s))

solution pH

2.9 39

0.6 101.0

4.82 f 4.14 ∼0

15

14.0

0.47 f 0.37

4.0 6.8 12.5 7.0 13 13 13 9.0

1.1 3.1 10.0 3.3 11.0 7.8 11.0 5.4

4.37 f 4.15 4.80 f 4.02 ∼2 ∼1 ∼0 ∼0 5.61 f 5.56

(8.0) (2.3) 1.6 1.38 0.5 0.4 0.125 negatived negativec,d

(4.3) (0.353) 0.17 0.13 0.0167 0.0107 0.001 -

∼4.0 2.38 f 2.41 5.57 f 5.39 9.42 f 9.09 ∼0 ∼0.07 9.56 f 9.49

a Nitrogen was used unless otherwise noted. Data inside parentheses are extrapolated values. “-” indicates no experimental data is available. b The second-order rate constant k2 was calculated from NHg ) (PHgi/HHg)xk2DHg-H2O[Hg(II)]i. c O2 contributed positively to Hg absorption. d Hgout > Hgin and Hgout . Hgin as HgCl2 was increased.

tion of a strong acid to Hg(II), such as HNO3 or H2SO4, greatly enhanced Hg0 absorption and changed the reaction to first-order in Hg2+. However, the addition of HCl inhibited Hg0 absorption in Hg(II).

Ind. Eng. Chem. Res., Vol. 37, No. 2, 1998 387

Succinic acid-NaOH buffer solution greatly enhanced Hg0 absorption by Hg(II) but NaHCO3-NaOH inhibited Hg absorption. Under most conditions, oxygen in the gas phase did not have any effect on Hg0 absorption by Hg(II). However, oxygen had a positive effect on Hg absorption by Hg(II) when HCl or NaHCO3-NaOH was present in the solution. The addition of MnSO4 to Hg(II) only slightly enhanced Hg absorption. On the other hand, NaCl, MgSO4, FeCl3, CaCl2, and MgCl2 all inhibited Hg absorption by Hg(II). When an oxidant was added to Hg(II), Hg0 absorption was further enhanced. The addition of H2O2 to HNO3 gave the most Hg0 removal. It was represented by kinetics that were first-order in H2O2 and Hg2+. The addition of Fe2+ or Fe3+ to Hg(II)-H2O2-HNO3 had no immediate effect. Both K2Cr2O7 and K2Cr2O7-HNO3 enhanced Hg0 absorption in Hg(II), and the positive effect of adding HNO3 was more apparent than that of adding K2Cr2O7. Acknowledgment This study was supported by Electric Power Research Institute Contract RP 3470-02. Notation [Cr2O72-]b ) Cr2O72- concentration in the bulk liquid phase (M) [Cr2O72-]i ) Cr2O72- concentration at the liquid side interface (M) DHg-H2O ) liquid film diffusion coefficient of Hg (m2 s-1) DHg2+-H2O ) liquid film diffusion coefficient of Hg2+ (m2 s-1) DHg-N2 ) interdiffusion coefficient of Hg and N2 (m2 s-1) HHg ) Henry’s constant of Hg in water (atm M-1) [H2O2] ) hydrogen peroxide concentration in the liquid (M) [H2O2]b ) hydrogen peroxide concentration in the bulk liquid phase (M) [H2O2]i ) hydrogen peroxide concentration at the liquid side interface (M) [Hg] ) Hg concentration in the liquid (M) [Hg(II)] ) Hg(II) concentration in the liquid (M) [Hg(II)]i ) Hg(II) concentration at the liquid side interface (M) [Hg2+]b ) Hg2+ concentration in the bulk liquid (M) [Hg2+]i ) Hg2+ concentration at the liquid side interface (M) k2 ) second-order rate constant (M-1 s-1) k3 ) third-order rate constant (M-2 s-1) k3′ ) third-order rate constant defined by eq 6 (M-2 s-1) kg ) gas film mass-transfer coefficient (mol s-1 atm-1 m-2) Kg′ ) normalized flux of Hg (NHg/PHgi) (mol s-1 atm-1 m-2) k°l,Hg ) physical liquid film mass-transfer coefficient of Hg (m s-1) k°l,Hg2+ ) physical liquid film mass-transfer coefficient of Hg2+ (m s-1)

[Mn(II)] ) Mn(II) concentration in the liquid (M) [Mn(II)]b ) Mn(II) concentration in the bulk liquid (M) [Mn(II)]i ) Mn(II) concentration at the liquid side interface (M) ng ) gas-phase agitation speed (rpm) nl ) liquid-phase agitation speed (rpm) NHg ) Hg flux (mol s-1 m-2) PHgb ) partial pressure of Hg in the bulk gas phase (atm) PHgi ) partial pressure of Hg at the gas side interface (atm) PHg,in ) partial pressure of Hg in the inlet gas (atm) T ) temperature (K)

Literature Cited Allgulin, T. Method of Purifying Gases Containing Mercury Vapor. U.S. Patent 3849537, 1974. Brosset, C. The Behavior of Mercury in the Physical Environment. Water, Air Soil Pollut. 1987, 34, 145. Clever, H. L.; Johnson, S. A.; Derrick, M. E. The Solubility of Mercury and Some Sparingly Soluble Mercury Salts in Water and Aqueous Electrolyte Solutions. J. Phys. Chem. Ref. Data 1985, 14, 631. Danckwerts, P. V. Gas-Liquid Reactions; McGraw-Hill Book Company: New York, 1970. Environmental Protection Agency, Code of Federal Regulations, 1992, title 40, part 266, App. IX, p 525. Kobayashi, T. Oxidation of Metallic Mercury in Aqueous Solutions by Hydrogen Peroxide and Chlorine. J. Jpn. Soc. Air Pollut. 1987, 22, 230. Lide, D. R., Ed. CRC Handbook of Chemistry and Physics, 75th ed.; CRC Press: Boca Raton, FL, 1994. Lindqvist, O.; Jernelov, A.; Johansson, K.; Rodhe, H. Mercury in the Swedish EnvironmentsGlobal and Local Sources. Report PM 1816; National Swedish Environmental Protection Board: Solna, Sweden, 1984. Morita, H.; Mitsuhashi, T.; Sakurai, H.; Shimomura, S. Absorption of Mercury by Solutions Containing Oxidants. Anal. Chim. Acta 1983, 153, 351. Munthe, J.; McElroy, W. J. Some Aqueous Reactions of Potential Importance in the Atmospheric Chemistry of Mercury. Atmos. Environ. 1992, 26A, 553. Parkinson, G., Ed. A Simplified Scrubbing Process Cuts the Cost of Flue Gas Cleanup. Chem. Eng. 1996, 6, 103. Peterson, J. R.; Hargrove O. W.; Seeger D. M.; Skarupa R. C.; Moser R. E. Mercury Emissions Control by Wet FGD Systems: EPRI Pilot-Scale Results. Proceedings of the 1995 SO2 Control Symposium, Miami, FL, March 28-31, 1995; Electric Power Research Institute: Palo Alto, CA, 1995. Sitaraman, R.; Ibrahim, S. H.; Kuloor, N. R. A Generalized Equation for Diffusion in Liquids. J. Chem. Eng. Data 1963, 8, 198. Zhao, L. L. Mercury Absorption in Aqueous Solutions. Ph.D. Dissertation, The University of Texas at Austin, Austin, TX, May, 1997. Zhao, L. L.; Rochelle, G. T. Hg Absorption in Aqueous Permanganate. AIChE J. 1996, 42, 3559.

Received for review February 10, 1997 Revised manuscript received December 1, 1997 Accepted December 1, 1997 IE970155O