Data Pack

Methods for the Determination of

Cyanide Using Amperometric Detection

June 2008
This Data Pack contains support data and apparatus information on three Lachat methods for the
determination of cyanide using amperometric detection. This allows for measurement of cyanide
without the use of hazardous chemicals like pyridine and barbituric acid, meeting a sizable demand
by customers in the Automated Photometry market.

Method Descriptions
Lachat QuikChem Method number 10-204-00-5-A utilizes Ligand Exchange reagents to free easily
liberated cyanides from methyl complexes. The range for this method is 2-400 μg CN-/L. This
method is equivalent to EPA method number OIA1677.

Lachat QuikChem Method number 10-204-00-5-B utilizes heat, acid, and UV light to digest cyanide
complexes in-line. The digested sample then passes through a diffusion block, where the cyanide
(now present as HCN(g)) is trapped in dilute NaOH. The cyanide present in the trapping solution is
separated utilizing gas diffusion, then measured amperometrically.

Lachat QuikChem method number 10-204-00-5-X allows samples distilled utilizing the MicroDIST
disposable distillation tubes to be measured through amperometric detection as well.

Special Apparatus
In addition to the QuikChem Flow Injection Analyzer and Method manifold, special apparatus is
required to use the Lachat QuikChem amperometric methods:

• Amperometric Detector apparatus (Lachat part no. 84920)

• Heating Unit (Lachat part no. A85100)
• Direct Voltage Detector (Lachat part no. 85272)

Users of the amperometric methods need a good understanding of the Omnion software, flow
injection analysis principles, and valve timing. Detailed instructions on how to install the required
apparatus and copies of the published methods are required for installation and successful analysis.
 QuikChem® Method 10-204-00-5-A

DETERMINATION OF AVAILABLE CYANIDE WITH

LIGAND DISPLACEMENT AND FLOW INJECTION
ANALYSIS (FIA) UTILIZING GAS DIFFUSION
SEPARATION AND AMPEROMETRIC DETECTION
2.0 to 400 μg CN-/L

– Principle –
Ligand-exchange reagents are added at room temperature to a cyanide-containing sample
in a pretreatment step. The ligand-exchange reagents form thermodynamically stable
complexes with the transition metal ions, resulting in the release of cyanide ion from the
metal-cyano complexes. The addition of hydrochloric acid converts cyanide ions to
hydrogen cyanide (HCN(g)) that passes through a gas diffusion membrane into an alkaline
receiving solution where it is converted back to cyanide ion. The cyanide ion is
monitored amperometrically with a silver working electrode, silver/silver chloride
reference electrode, and platinum/stainless steel counter electrode, at an applied potential
of zero volts. The current generated is proportional to the cyanide concentration present
in the original sample.

– Interferences –
1. High levels of carbonate can release CO2 into the acceptor stream and cause an
interference with the amperometric detector that results in a slight masking effect
(15 percent negative bias with 20 ppb cyanide in 1500 ppm carbonate).
2. Sulfide will diffuse through the gas diffusion membrane and can be detected in
the amperometric flowcell. Oxidized products of sulfide can also rapidly convert
CN- to SCN- at a high pH.
3. Refer to Section 4 of this method for additional information regarding interferents
in the analysis of cyanide.

Amperometric methods 02 June 2008

Calibration Data for Amperometric Cyanide

File Name: 4-30 cal support.OMN

Acq. Date: 30 April 2008

Calibration Graph and Statistics

Amperometric methods 2 02 June 2008

Method Detection Limit for cyanide using a 1.0 μg CN-/L standard
MDL= 0.65 μg CN-/L
Standard Deviation (s) = 0.208 μg CN-/L, Mean (x) = 1.28 μg CN-/L, Known value = 1.0 μg CN-/L
File Name: 4-30 cal support.OMN
Acq. Date: 30 April 2008

Precision data for cyanide using a 100 μg CN-/L standard

% RSD = 1.16 %
Standard Deviation (s) = 1.13 μg CN-/L, Mean (x) = 97.32 μg CN-/L, Known value = 100 μg CN-/L
File Name: 4-30 cal support.OMN
Acq. Date: 30 April 2008

Amperometric methods 3 02 June 2008

Carryover Study: 400 μg CN-/L standard followed by 3 blanks
Carryover Passed
File Name: 4-30 CO DIN.omn
Acq. Date: 30 April 2008

DIN Blanks

Amperometric methods 4 02 June 2008

Average: -0.037 μg CN-/L, SD = 0.166 μg CN-/L. Calculated DIN Limits: Detection
Limit = 0.498 μg CN-/L, Decision Limit = 0.996 μg CN-/L, Determination Limit = 1.49
μg CN-/L;
File Name: 4-30 CO DIN.omn
Acq. Date: 30 April 2008

Amperometric methods 5 02 June 2008

Recovery of Nickel cyanide and Mercury cyanide complexes

Compound Amount in Sample Average Value Obtained % Recovery *

Nickel cyanide 100.0 μg CN-/L 95.64 μg CN-/L 95.64
Mercury cyanide 100.0 μg CN-/L 103.6 μg CN-/L 103.6
*(determined/known) * 100
Conclusion: Nickel and Mercury cyanide are recovered at levels greater than 95%.

Cyanide Spike Recoveries in 5-Mile Effluent

Sample ID Average spike recoveries Spike Level % Recovery

(μg CN-/L) (μg CN-/L)
5-M effluent 4.25 100 ---
5-M KCN 106.5 100 102.25
5-M K2Ni(CN)4 92.92 100 88.67

Amperometric methods 6 02 June 2008

 5-M Hg(CN)2 96.00 100 91.75
Conclusion: Potassium, Nickel and Mercury cyanide are recovered at levels greater than 88%.

Cyanide Spike Recoveries in California Tap Water

Sample ID Average spike recoveries Spike Level % Recovery

(μg CN-/L) (μg CN-/L)
CA Tap Water -0.176 100 ---
CA Tap Water KCN 102.2 100 102.4
CA Tap Water K2Ni(CN)4 94.75 100 94.75
CA Tap Water Hg(CN)2 97.31 100 97.31
Conclusion: Potassium, Nickel and Mercury cyanide are recovered at levels greater than 94%.

Cyanide Spike Recoveries in Loveland, CO Tap Water

Sample ID Average spike recoveries Spike Level % Recovery

(μg CN-/L) (μg CN-/L)

Amperometric methods 7 02 June 2008

Loveland Tap Water -0.368 100 ---
Loveland Tap Water KCN 104.4 100 104.8
Loveland Tap Water K2Ni(CN)4 92.94 100 93.31
Loveland Tap Water Hg(CN)2 98.65 100 99.02
Conclusion: Potassium, Nickel and Mercury cyanide are recovered at levels greater than 93%.

Cyanide Spike Recoveries in 0.025 N NaOH

Sample ID Average spike recoveries Spike Level % Recovery

(μg CN-/L) (μg CN-/L)
0.025 N NaOH -0.271 100 ---
0.025 N NaOH KCN 92.22 100 92.22
0.025 N NaOH K2Ni(CN)4 95.18 100 95.46
0.025 N NaOH Hg(CN)2 101.6 100 101.9
Conclusion: Potassium, Nickel and Mercury cyanide are recovered at levels greater than 92%.

Amperometric methods 8 02 June 2008

Spike Recoveries from Previously Reported Data

Sample Sample EXP. Results *Results **Recovery

ID (spike level 94.34 μg CN/L) (μg CN/L) (μg CN/L) (%)
bl-cn KCN spike into blank 92.62 94.1 100
bl-nicn (K2Ni(CN)4 H2O) spike into blank 93.17 94.7 100.6
bl-hgcn (Hg(CN)2) spike into blank 94.75 96.3 102.3
bl Blank (0.025M NaOH solution) -1.52
ete-cn KCN spike into ETE sample 92.62 94.0 99.9
ete-nicn (K2Ni(CN)4 H2O) spike into ETE sample 98.17 99.6 105.8
ete-hgcn (Hg(CN)2) spike into ETE sample 94.37 95.8 101.7
ete ETE sample -1.40
wte-cn KCN spike into WTE sample 89.86 90.9 96.5
wte-nicn (K2Ni(CN)4 H2O) spike into WTE sample 96.89 97.9 104.0
wtehgcn (Hg(CN)2) spike into WTE sample 92.07 93.1 98.9
wte WTE sample -1.01
ste-cn KCN spike into STE sample 94.59 94.5 100.4
ste-nicn (K2Ni(CN)4 H2O) spike into STE sample 98.11 98.1 104.2
ste-hgcn (Hg(CN)2) spike into STE sample 92.35 92.3 98.0
ste STE sample 0.05
* Results = (EXP results – Blank)
** Recovery: Assume that the simple cyanide in blank are recovered 100%.
Conclusion:
1) Ligand exchange solution is working. Both nickel cyanide and mercury cyanide in
blank are recovered 100%.
2) There is no sample matrix effect. All of the simple and complex cyanide tested were
recovered at greater than 95%.

Amperometric methods 9 02 June 2008

 QuikChem® Method 10-204-00-5-B

DETERMINATION OF TOTAL CYANIDE BY FLOW

INJECTION ANALYSIS UTILIZING IN-LINE DIGESTION,
GAS DIFFUSION SEPARATION AND AMPEROMETRIC
DETECTION

2.0 to 500 μg CN-/L

– Principle –
Liquid samples are first mixed with phosphoric acid, and then introduced into an in-line
cyanide distillation unit. In this unit, the samples are first heated to 135 oC and then
passed over a Black-Light Blue UV lamp (radiation peaks at 352 nm and 368 nm) to
cleave metal-CN complexes. The CN- released from these complexes during the heating
and irradiation combines with protons to form HCN(g). The sample then exits the
distillation unit and passes through a diffusion block where the HCN(g) diffuses across a
Teflon membrane and is trapped as CN- in a sodium hydroxide solution flowing across
the bottom of the membrane (note: pKa of HCN is ~ 9.2-9.3). The cyanide ion is
monitored amperometrically with a silver working electrode, silver/silver chloride
reference electrode, and platinum/stainless steel counter electrode, at an applied potential
of zero volts. The current generated is proportional to the cyanide concentration present
in the original sample.

– Interferences –
1. High levels of carbonate can release CO2 into the acceptor stream and cause an
interference with the amperometric detector that results in a slight masking effect
(15 percent negative bias with 20 ppb cyanide in 1500 ppm carbonate).
2. Sulfide will diffuse through the gas diffusion membrane and can be detected in
the amperometric flowcell. Oxidized products of sulfide can also rapidly convert
CN- to SCN- at a high pH.
3. Refer to Section 4 of this method for additional information regarding interferents
in the analysis of cyanide.
-
4. Thiocyanate is not a significant interference in this method. A 100 μg/L SCN standard
- -
gave an average CN response of –1.12 μg CN /L.

Amperometric methods 10 02 June 2008

Calibration Data for Total Cyanide

File Name: 5-21 cal support.omn

Acq. Date: 21 May 2008

Calibration Graph and Statistics

Amperometric methods 11 02 June 2008

Amperometric methods 12 02 June 2008
Method Detection Limit for cyanide using a 1.0 μg CN-/L standard
MDL= 0.234 μg CN-/L, Reporting 0.914 μg CN-/L due to carry over.
Standard Deviation (s) = 0.074 μg CN-/L, Mean (x) = 1.08 μg CN-/L, Known value = 1.0 μg CN-/L
File Name: 5-22 mdl.omn
Acq. Date: 22 May 2008

Precision data for cyanide using a 100 μg CN-/L standard

% RSD = 0.96 %
Standard Deviation (s) = 0.94 μg CN-/L, Mean (x) = 97.72 μg CN-/L, Known value = 100 μg CN-/L

Amperometric methods 13 02 June 2008

File Name: 5-21 cal support.omn
Acq. Date: 21 May 2008

Carryover Study: 500 μg CN-/L standard followed by 3 blanks

Carryover Failed, Reporting MDL as 0.914 μg CN-/L
File Name: 5-22 CO.omn
Acq. Date: 22 May 2008

DIN Blanks
Average: -1.512 μg CN-/L, SD = 0.270 μg CN-/L. Calculated DIN Limits: Detection
Limit = 0.811 μg CN-/L, Decision Limit = 1.623 μg CN-/L, Determination Limit = 2.434
μg CN-/L;
File Name: 5-22 DIN.omn
Acq. Date: 22 May 2008

Amperometric methods 14 02 June 2008

Recovery of Ferro, Ferri and Nickel cyanide complexes

Compound Amount in Sample Average Value Obtained % Recovery *

Ferro cyanide 100.0 μg CN-/L 104.95 μg CN-/L 104.95
Ferri cyanide 100.0 μg CN-/L 98.67 μg CN-/L 98.67
Nickel cyanide 100.0 μg CN-/L 100.25 μg CN-/L 100.25
*(determined/known) * 100
Conclusion: Ferro, Ferri, and Nickel cyanide are recovered at levels greater than 98%.

Amperometric methods 15 02 June 2008

Cyanide Spike Recoveries in 5-Mile Effluent

Sample ID Average spike recoveries Spike Level % Recovery

(μg CN-/L) (μg CN-/L)
5-M effluent 25.92 100 ---
5-M KCN 128.6 100 102.7
5-M Ferro-CN 121.6 100 95.63
5-M Ferri-CN 115.8 100 89.88
Conclusion: Potassium, Ferro, and Ferri cyanide are recovered at levels greater than 90%. Nickle cyanide
only recovered at 50% in the effluent sample.

Amperometric methods 16 02 June 2008

 Cyanide Spike Recoveries in Loveland, CO Tap Water

Sample ID Average spike recoveries Spike Level % Recovery

(μg CN-/L) (μg CN-/L)
Loveland Tap Water -1.90 100 ---
Loveland Tap Water KCN 89.19 100 91.09
Loveland Tap Water Ferro-CN 101.4 100 103.4
Loveland Tap Water Ferri-CN 94.32 100 96.22
Loveland Tap Water K2Ni(CN)4 100.8 100 102.7
Conclusion: Potassium, Ferro, Ferri, and Nickel cyanide are recovered at levels greater than 91%.

Amperometric methods 17 02 June 2008

Interference due to Thiocyanate (SCN-)

Recovery of SCN- at the beginning of the run, average recovery –0.28 μg CN-/L (SCN- as CN-)

Recovery of SCN- at the end of the run, average recovery –1.97 μg CN-/L (SCN- as CN-)
Conclusion: The interference from Thiocyanate is negligible, with an average of –1.12 μg CN-/L
throughout the course of the analytical run. It is suggested that during the analytical run the analyst inject a
100 ppb SCN- standard to ensure that the membrane has retained its hydrophobic nature, see section 11.3.4
for more information.

Amperometric methods 18 02 June 2008

 QuikChem® Method 10-204-00-5-X

Cyanide in Drinking and Wastewaters

using the MICRO DISTTM Distillation
and Utilizing Gas Diffusion Separation
and Amperometric Detection
5.0 to 400 μg CN-/L

– Principle –
By means of a passive miniature distillation device, MICRO DIST, the cyanide in the
samples is released by digesting and acidifying cyanide complexes, converting them to
hydrocyanic acid (HCN). The cyanide ion is trapped in a 0.1 M sodium hydroxide
absorbing solution which is diluted to 0.025 M solution during the distillation. By means
of flow injection, the hydrocyanic acid (HCN) passes through a gas diffusion membrane
into an alkaline receiving solution where it is converted back to cyanide ion. The cyanide
ion is monitored amperometrically with a silver working electrode, silver/silver chloride
reference electrode, and platinum/stainless steel counter electrode, at an applied potential
of zero volts. The current generated is proportional to the cyanide concentration present
in the original sample.

– Interferences –
1. Most non-volatile interferences are eliminated or minimized by the distillation
procedure. Some of the known interferences are aldehydes, nitrate-nitrite, and
oxidizing agents, such as chlorine, thiosulfate, and sulfide. Multiple interferences
may require the analysis of a series of laboratory fortified sample matrices (LFM)
to verify the suitability of the chosen treatment.
2. Thiocyanate will interfere if present. This method should not be used if sample
thiocyanate concentrations exceed 0.002 mg/L. Instead, a method based on weak
acid dissociable, or ligand-exchange digestion should be used.
3. Sulfides adversely affect the procedure by producing hydrogen sulfide during
distillation. If a drop of the sample on lad acetate test paper indicates the presence
of sulfide, treat 25 mL more than the stabilized sample (pH ≥ 12) than that
required for the cyanide determination with powdered cadmium carbonate.
Yellow cadmium sulfide precipitates if the sample contains sulfide. Repeat this
operation until a drop of the treated sample solution does not darken the lead
acetate test paper. Filter the solution through a dry filter paper into a dry beaker,
and from the filtrate, measure the sample to be used for analysis. Avoid a large

Amperometric methods 19 02 June 2008

 excess of cadmium and a long contact time in order to minimize a loss by
complexation or occlusion of cyanide on the precipitated material.
4. Studies have shown that sulfide concentrations of up to 10 mg S2-/L in the
distillate can be tolerated. However, when is it expected that hydrogen sulfide
will be generated from the distilled sample during the distillation, every effort
should be made to analyze distillates within 2 hours from the start of distillation.
Sulfide will diffuse through the gas diffusion membrane and can be detected in
the amperometric flowcell. Oxidized products of sulfide can also rapidly convert
CN- to SCN- at a high pH.
5. If exceptionally high concentrations of sulfide (>> 10 mg S2-/L) are generated
during distillation they could significantly bias results, violating the QC criteria in
Section 9 of this method. Should this situation occur, the distillate should be
treated with cadmium carbonate and filtered per Section 11.1.1, or a MICRO
DIST tube containing the lead cation, as described in other approved distillation
methods, should be used. When using a sulfide removal procedure, make sure to
initially analyze QC samples to ensure criteria are being met with the modified
procedure.
Note: If samples contain particulate that would be removed upon filtration, the
samples must be filtered prior to treatment with cadmium carbonate. The
collected particulate must be saved, and the filtrate then treated using the sulfide
removal procedure above. The collected particulate and treated filtrate must be
recombined, homogenized, and included in the total cyanide distillation.
6. High results may be obtained for samples that contain nitrate and/or nitrite.
During the distillation nitrate and nitrite will form nitrous acid that will react with
some organic compounds to form oximes. These oximes will decompose under
test conditions to generate HCN. The interference of nitrate and nitrite is
eliminated by pretreatment with sulfamic acid.
7. Oxidizing agents, such as residual chlorine, decompose most of the cyanides.
Test a drop of the sample with potassium iodide (KI)-starch paper at time of
collection; a blue color indicates the need for treatment. Add ascorbic acid, a few
crystals (about 0.6 g each) at a time, until a drop of sample produces no color in
the indicator paper; then add an additional 0.06h g of ascorbic acid for each liter
of sample volume. Sodium arsenite has also been employed to remove oxidizing
agents.
8. High levels of carbonate can release CO2 into the acceptor stream and cause an
interference with the amperometric detector that results in a slight masking effect
(15 percent negative bias with 20 ppb cyanide in 1500 ppm carbonate).
9. Method interferences may be caused by contaminants in the reagent water,
reagents, and sample processing apparatus that bias response.

Amperometric methods 20 02 June 2008

Calibration Data for Cyanide (MicroDistTM Distilled Standards)

File Name: 5-22 Cal spikes.omn

Acq. Date: 22 May 2008

Calibration Graph and Statistics (Distilled Standards)

Amperometric methods 21 02 June 2008

Calibration Data for Cyanide (Non-Distilled Standards)

File Name: 5-9 cal support no dist.OMN

Acq. Date: 9 May 2008

Calibration Graph and Statistics (Non-Distilled Standards)

Amperometric methods 22 02 June 2008

Method Detection Limit for cyanide using a 4.0 μg CN-/L standard (Distilled Standards)
MDL= 0.975 μg CN-/L
Standard Deviation (s) = 0.346 μg CN-/L, Mean (x) = 4.44 μg CN-/L, Known value = 4.0 μg CN-/L
File Name: 5-22 Prec mdl.omn
Acq. Date: 22 May 2008

Precision data for cyanide using a 100 μg CN-/L standard (Distilled Standards)
% RSD = 1.53 %
Standard Deviation (s) = 1.56 μg CN-/L, Mean (x) = 101.75 μg CN-/L, Known value = 100 μg CN-/L
File Name: 5-22 Prec mdl.omn
Acq. Date: 22 May 2008

Amperometric methods 23 02 June 2008

Method Detection Limit for cyanide using a 1.0 μg CN-/L standard (Non-Distilled
Standards)
MDL= 0.20 μg CN-/L
Standard Deviation (s) = 0.064 μg CN-/L, Mean (x) = 1.10 μg CN-/L, Known value = 1.0 μg CN-/L
File Name: 5-8 cal mdl no dist.OMN
Acq. Date: 8 May 2008

Precision data for cyanide using a 50 μg CN-/L standard (Non-Distilled Standards)

% RSD = 0.56 %
Standard Deviation (s) = 0.27 μg CN-/L, Mean (x) = 48.49 μg CN-/L, Known value = 50 μg CN-/L
File Name: 5-9 cal support no dist.OMN
Acq. Date: 9 May 2008

Amperometric methods 24 02 June 2008

Carryover Study: 400 μg CN-/L standard followed by 3 blanks (Non-Distilled Standards)
Carryover Passed
File Name: 5-9 cal support no dist.OMN
Acq. Date: 9 May 2008

DIN Blanks (Non-Distilled Standards)

Average: -0.037 μg CN-/L, SD = 0.166 μg CN-/L. Calculated DIN Limits: Detection
Limit = 0.498 μg CN-/L, Decision Limit = 0.996 μg CN-/L, Determination Limit = 1.49
μg CN-/L;
File Name: 4-30 CO DIN.omn
Acq. Date: 30 April 2008

Amperometric methods 25 02 June 2008

 Cyanide Spike Recoveries in Loveland, CO Tap Water

Sample ID Average spike recoveries Spike Level % Recovery

(μg CN-/L) (μg CN-/L)
Loveland Tap Water 0.256 100 ---
Loveland Tap Water KCN 97.12 100 96.86
Loveland Tap Water Ferro-CN 120.1 100 119.8
Loveland Tap Water Ferri-CN 108.9 100 108.6
Loveland Tap Water K2Ni(CN)4 106.1 100 105.8
Conclusion: Potassium, Ferro, Ferri and Nickel cyanide are recovered at levels greater than 96%.

Amperometric methods 26 02 June 2008

Cyanide Spike Recoveries in 5-Mile Effluent

Sample ID Average spike recoveries Spike Level % Recovery

(μg CN-/L) (μg CN-/L)
5-M effluent 56.40 --- ---
5-M KCN 149.1 100 92.66
5-M Ferro-CN 179.2 100 122.8
5-M Ferri-CN 166.3 100 109.9
5-M K2Ni(CN)4 136.8 100 80.41
5-M Hg(CN)2 149.2 100 92.81
Conclusion: Potassium, Ferro, Ferri, Nickel and Mercury cyanide are recovered at levels greater than
80%.

Amperometric methods 27 02 June 2008

Recovery of Nickel cyanide and Mercury cyanide complexes

Compound Amount in Sample Average Value Obtained % Recovery *

Nickel cyanide 100.0 μg CN-/L 97.81 μg CN-/L 97.81
Mercury cyanide 100.0 μg CN-/L 96.94 μg CN-/L 96.84
*(determined/known) * 100
Conclusion: Nickel and Mercury cyanide are recovered at levels greater than 96%.

Amperometric methods 28 02 June 2008