An NPDES Distillation Method for Measuring Total Cyanide in Water

Used for total cyanide distillation, the MICRO DIST system (Lachat Instruments-Hach, Loveland, CO) was granted National Primary Drinking Water Regulations (NPDWR) approval in 2003 and is expected to receive nationwide Tier III approval for National Pollution Discharge Elimination System (NPDES) in October 2005.1 The MICRO DIST (Figure 1) can be used instead of the cumbersome, slow, and error-prone large-scale glassware apparatus described in U.S. EPA Method 335.22 and Standard Methods 4500-CN-C.3 This study evaluates the proposed distillation for total cyanide in a variety of water matrices with two flow injection analysis systems as the determinative step.

Figure 1 - Samples and standards distilling simultaneously in a MICRO DIST block.

As part of the study, interlaboratory and intralaboratory tests were performed to evaluate method performance. The method detection limit, precision, and recoveries of selected metal cyanide complexes were evaluated. Samples containing Prussian blue were distilled with the MICRO DIST system and compared with the U.S. EPA distillation to evaluate the recovery of colloidal cyanides. The sample pH, storage temperature, holding time, and interference issues were also investigated.

Experimental

Standards and environmental samples were digested and distilled using the MICRO DIST system and were evaluated with a QuikChem automated ion analyzer (Lachat Instruments-Hach) or a flow injection analysis (FIA) instrument equipped with gas diffusion separation and amperometric detection. Samples (6 mL each) were pipetted into a sample tube and then acidified with 0.75 mL of magnesium chloride/sulfuric acid-releasing solution. The samples were distilled for 30 min at 120 °C in a special heating block that fits the sample tube closely, allowing the poorly conducting polypropylene to be heated rapidly yet locally. As the sample boiled the vapors passed through a hydrophobic, porous membrane. The vapors condensed above the membrane to form a liquid pool over the membrane that could not pass back through into the sample.

A 1.0 M sodium hydroxide trapping solution was added above the membrane to convert hydrogen cyanide gas to the nonvolatile cyanide anion. After the distillation, the top of each tube was separated from the bottom at the breakaway point (Figure 2) and the distillate sample was diluted to the original volume, equivalent to 0.25 M NaOH, prior to analysis. The cyanide in the distillates was analyzed with spectrophotometric or amperometric detection systems. The two determinative steps for the distillates are described below.

Figure 2 - MICRO DIST tube.

For spectrophotometric determinations, the cyanide present in the 0.25 M NaOH distillate was converted to cyanogen chloride by reaction with chloramine-T, pyridine, and barbituric acid to provide a red-colored complex. The absorbance of this complex was measured at 570 nm by measuring the peak area resulting from the sample. The peak area is proportional to the concentration of the cyanide in the sample.

For amperometric determinations, the distilled samples were analyzed as described in ASTM Standard Test Method D 6888- 04.4 In this method, the distillates are injected into an FIA system and acidified on-line with dilute sulfuric acid and bismuth nitrate. The cyanide diffuses through a gas diffusion membrane and is captured into an alkaline acceptor stream. The captured cyanide is sent to an amperometric flow cell detector where the measured anodic current is proportional to the cyanide concentration.

Effect of method detection limit

The method detection limit (MDL) for each determinative step was established according to the U.S. EPA procedure.5 The MDL is defined by the U.S. EPA as “the minimum concentration of a substance that can be measured and reported with 99% confidence that the analyte concentration is greater than zero and is determined from analysis of a sample in a given matrix containing the analyte.”

A 5-μg/L CN standard was distilled seven times and analyzed by the spectrophotometric method shown in Figure 3. A 10-μg/L CN standard was distilled seven times and analyzed by the amperometric method. The calculated MDL values were <2 μg/L CN for both determinative steps, which is at least 10 times lower than the detection limit published in U.S. EPA Method 335.2.

Figure 3 - Method detection limit data for total cyanide with 5 μg/L CN.

Recoveries of potassium ferricyanide [K3Fe(CN)6]

In the total cyanide method, many metal complexes of cyanide are recovered. As a recovery test, laboratory water samples were fortified with K3Fe(CN)6 at 300 μg/L CN or 500 μg/L CN. The metal cyanide complex of [Fe(CN)6]3– was chosen because of its relatively high stability constant (43.6 log K at 25 °C) in order to challenge the distillation.6 The mean recovery was 305 μg/L CN or 99.8%, with an RSD of 1.5% (spectrophotometric detection, n = 5) and a mean recovery of 503 μg/L CN or 101%, with an RSD of 1.5% (amperometric detection, n = 4).

Three wastewater samples were fortified with K3Fe(CN)6, distilled with the MICRO DIST system, and then analyzed with amperometric detection to evaluate recovery. The samples were believed to have a sulfide interference with the U.S. EPA method; however, the spike recoveries were found to be acceptable with the evaluated procedure. The data demonstrate that sulfide can be mitigated with the bismuth nitrate that is in the acidification solution of ASTM D 6888-04 (see Table 1).

Table 1 - Fortified wastewater*

Quality control samples

A reference water sample from Environmental Resource Associates (ERA, Arvada, CO), lot no. 9959, was digested and distilled in duplicate. Its certified concentration for total cyanide was 219 μg/L CN. The observed concentration with the spectrophotometric determination was 195 μg/L CN, which resulted in an 88.9% recovery. Quality control data are tabulated in Table 2 for samples that were distilled with the MICRO DIST system and then analyzed by amperometric detection. All of the quality control samples were within the acceptance range for each QC sample.

Table 2 - Quality control summary for MICRO DIST followed by amperometric detection

Tier III alternative test procedure (ATP) matrices and recovery results

In an effort to obtain nationwide approval for drinking water and wastewater, a Tier III ATP was conducted. This process required that nine laboratories submit recovery and performance data for nine separate matrices. The recoveries for matrix spikes and the relative percent difference on their duplicates are shown in Table 3. This study was conducted using the MICRO DIST system followed by spectrophotometric determinations.

Table 3 - U.S. EPA Tier III summary

Prussian blue recoveries

A Prussian blue recovery solution was prepared by accurately weighing 11.0 mg of the pigment, ferri-ferrocyanide (CAS 14038-43-8), into 1 L total volume 0.01 M sodium hydroxide, resulting in a theoretical concentration of 6.00 mg/L as CN. When Prussian blue is subjected to alkaline conditions of sample preservation, it forms a brown-colored solution of iron(III) hydroxide and ferrocyanide that should return to blue upon acidification. Tests were conducted to compare the cyanide recoveries of the MICRO DIST system with a traditional midi-distillation.

The stock solution of Prussian blue was diluted to 300 μg/L as CN in 0.01 M sodium hydroxide (all of the Prussian blue appeared to dissolve), and 10 replicates were distilled with each method for total cyanide. All distillates were analyzed with ASTM D 6888-04 (gas diffusion separation and amperometric detection). The data are summarized in Table 4. In addition, a manufacturing wastewater was fortified with Prussian blue with a mean recovery of 82.2% using the MICRO DIST system.

Table 4 -Prussian blue recovery study

During the study it was discovered that temperature, pH, and storage time affected the recovery of Prussian blue. Samples subjected to longer storage time at pH 12 showed significantly improved recoveries. If samples are suspected to contain Prussian blue, it is imperative to preserve the samples with sodium hydroxide to pH 12 upon sample collection and allow the samples sufficient time to reach room temperature prior to analysis. For example, the recovery of cyanide from Prussian blue in a sample that was stored at 4 °C increased from 64% to 84% by allowing the sample to warm to room temperature prior to analysis.

Interference issues

Interference issues from sulfide (S2–), nitrates/nitrites (NOx), and thiocyanate (SCN) could potentially cause false positive or negative results if not mitigated.7 Sulfide is typically removed by adding lead carbonate to the sample and/or to the absorbing solution during the distillation; however, the rapid loss of cyanide is possible due to the formation of thiocyanate.8 The data from this study demonstrate that sulfide can be removed by performing the determinative step employing the sulfide abatement procedure described in ASTM D 6888-04 using bismuth nitrate in the acidification reagent during flow injection analysis followed by amperometric detection. For spectrophotometric determinations, lead carbonate can be added to the absorbing solution after the distillation is complete, followed by immediate filtration to avoid the formation of thiocyanate.

Nitrate and nitrite in the presence of thiocyanate or other compounds may form HCN during the distillation. This can be avoided by adding 0.25 mL 1 M sulfamic acid solution to the sample during the distillation step. However, up to 50 μg/L CNpositive interference was observed in synthetic precious metals mining wastewater samples with 15 mg/L SCN in the presence of 25 mg/L NH3 as N and 25 mg/L NO3 as N. This interference is likely due to the harsh conditions of the distillation (high temperature and acidic conditions) and also occurs with U.S. EPA Method 335.2; therefore, it is not recommended to use any distillation method with this particular sample type.

Conclusion

The MICRO DIST system proved to be a good alternative to conventional cyanide distillations. The system uses 80 times less sample, yet is capable of a detection limit that is 10 times lower than that of U.S. EPA Method 335.2. The system takes considerably less time to set up, and throughput of samples is greatly increased to approx. 21 distillations per hour. Since the MICRO DIST tubes are single use, there is essentially little or no cleanup. Recoveries were adequate for all total cyanide species evaluated during the study. The method is applicable to a wide range of sample matrices, including samples that contain high levels of suspended solids or colloidal cyanides. Interference issues can be mitigated in most samples using the techniques described in this article.

References

  1. Fed Reg Apr 6, 2004. Proposed rules, vol 69, no. 66.
  2. U.S. EPA Method 335.2. Cyanide Total, 1980.
  3. Method 4500-CN-C. Total Cyanide After Distillation. Standard Methods for the Examination of Water and Wastewater, 20 ed., APHA, 1998.
  4. ASTM D 6888-04. Standard Test Method for Available Cyanide With Ligand Displacement and Flow Injection Analysis Utilizing Gas Diffusion Separation and Amperometric Detection, 2004.
  5. Appendix B, 40 CFR Part 136. Definition and Procedure for the Determination of the Method Detection Limit, rev. 1.11, 1986.
  6. ASTM D 6696-01. Standard Guide for Understanding Cyanide Species, 2001.
  7. ASTM D 2036-98. Standard Test Method for Cyanides in Water, 1998.
  8. Wilmot JC, Solujic L, et al. Formation of thiocyanate during removal of sulfide as lead sulfide prior to cyanide determination. Analyst Jun 1996; 121:799–801.

Mr. Sebroski is Associate Scientist, Materials Characterization, Bayer Material Science LLC, 100 Bayer Rd., Pittsburgh, PA 15205, U.S.A.; tel.: 412-777-3420; fax: 412-777-7640, e-mail: [email protected]. Ms. Bogren is Program Manager, Lachat Instruments-Hach, 5600 Lindbergh Dr., Loveland, CO 80538, U.S.A.; tel.: 970-663-1377; fax: 970-962-5610; e-mail: [email protected].

Comments