CENTRO DE ENSEANZA TCNICA Y SUPERIOR

Engineering School
DOE
"Cyanide recovering & regeneration
From industrial effluents (gold & silver metals extraction process)
Samantha Pez 14633

Tijuana B.C. October 1, 2015.

CYANIDE RECOVERY TECHNOLOGY COMPARISON1

Legislation in most gold mining regions of the world demands that gold plants meet rigorous
environmental standards with respect to cyanide. These requirements can be met either by destroying
cyanide after a single pass through the gold plant or by recovering cyanide from the tailings and recycling it
to leaching. Although cyanide recovery processes are not yet widely practiced, the technologies are fairly
simple, both from chemistry and process engineering perspectives, and they have the potential to yield
significant economic and environmental benefits.
AVAILABLE TECHNOLOGIES FOR CYANIDE RECOVERY
There are several processes for recovering cyanide from gold plant tailings that are available to the
mining industry.
1. Cyanisorb (AVR: Acidify, volatilize and reneutralize)
2. SART (Sulphidize, acidify, recycle and thicken)
3. Hannah (Strong base resin extraction of free cyanide and metal cyanide species)
4. AuGMENT (Strong base resin extraction of copper and cyanide). SGS metallurgical group has been
involved in the development of all of these processes and can determine the advantages and
disadvantages of each.
COST EFFECTIVENESS IS CRITICAL
The cost of cyanide consumption and destruction can be a significant percentage of total operating
costs on a gold plant, and unlike other operating costs, yields no economic return. The choice of the best
strategy for handling cyanide in the tailings can therefore have a major influence on the economic health of
a project. The economic impact of cyanide is different in every operation, but it is possible to compare cyanide
destruction and cyanide recovery costs for a fairly generic high cyanide consuming case. The capital cost of
a cyanide recovery plant could pay back in less than 2 years in many cases, particularly when high cyanide
consumption is associated with copper mineralization in the ore. As a broad generalization, cyanide recovery
will produce more favorable economics than cyanide destruction in those cases where potentially recoverable
cyanide consumption is greater than 1 kg/t NaCN, or exceeds 1000 tons NaCN per annum.
WHY RECYCLE CYANIDE? LOWER COSTS
Recycled cyanide is 2-3 times cheaper than new cyanide.
Reduced detoxification costs.
Opportunity to sell by-products such as copper sulphide.
LESS ENVIRONMENTAL RISK
Reduced loading of potentially harmful salts in the tailings and recycle water (e.g. CNS, CNO, NH 3 and
metal cyanides).
Less new cyanide is needed, so there is less transported, stored and handled, reducing the risk of spills.

CYANIDE RECOVERY: http://www.sgs.mx/~/media/Global/Documents/Flyers%20and%20Leaflets/SGS-MINWA016-Cyanide-Recovery-Comparison-EN-11.pdf

So as we learned Cyanide ion (CN-) is used for precious (Gold & Silver) metals extraction process
because CN- make very stable compounds in solution. The disadvantage is CN- has a very high toxic levels,
so it is important recover the CN- & avoid impact the environment within effluents.
Recover it or "reuse" (regeneration) are options to avoid effluents with high levels of CN-.
Next data come from experimental work using pressured air through porous tube. Air - porous tube
generate very fine bubbles which extract the CN- as gas from the solution (effluent) & release the effluent
from the CN-.
After that the CN- recovered into the experiment, can be used again to extract metals.
The data table contains:
1. CN- solution flow (liter / minute)
2. Rate (proportion) air versus solution.
NOTE: Is expected higher air proportion means higher recovering rate means Lower CN- concentration into
effluent.
Cyanide (CN-) remaining within effluent (CN- not recovered) (5 repetitions)
The objective for the experiment is demonstrates if there is some relation between PROPORTION AirSolution (FACTOR) & CN- not recovered (RESPONSE)

Initial Concentration for

CNAir flow

CN- solution flow

(liters/min)
50
40
30
20

250

mg / liters

210

Liters/min

CN- Not recovered present within effluent (mg/liters)

PROPORTION
Rep 1
Rep 2
Rep 3
Rep 4
Rep 5
Air / Solution
4.20
172
202
176
190
181
5.25
120
109
104
125
116
7.00
94
86
105
100
96
10.50
39
66
49
48
50
Table 1. Data received from the experiment

1. Give both hypothesis (name & formula).

2. Calculate % of recovery for each treatment. % Recovering = (Initial Concentration - Conc. Not
recovered)/Initial Conc.
3. Obtain chart % of recovering versus Proportion air-solution.
a. Indicate what kind of relation is there? (Direct / Reverse; linear, nonlinear, etc.
b. What happen with % of recovering if we use more air? Increase or decrease?
4. Apply all methods recommended by MINITAB using ANOVA to detect if there is statistical difference
a. FACTOR: PROPORTION AIR SOLUTION
b. RESPONSE: Concentration not recovered
5. Built on CHARTS with AVERAGE & BLOCKS to confirm if there is difference (No overlap)
6. Conclusions. Differences between each pair compared
7. Obtain 4 in 1 charts for RESIDUALS
8. Analyze R2 for linear model

EXPERIMENT
Hypothesis: Given the last information is expected to higher air proportion, higher recovering rate and lower
CN- concentration into effluent. In simple words, to more pressure given, more cyanide recover.

RESULTS
Air/Solution Proportion
(liter/minute)

Not Recovery Cyanide

(mg/liter)

FITS

RESI

FITS_1

RESI_1

4.20

172

184.20

184.20

-12.20

4.20
4.20
4.20
4.20
5.25
5.25
5.25

202
176
190
181
120
109
104

184.20
184.20
184.20
184.20
114.80
114.80
114.80

184.20
184.20
184.20
184.20
114.80
114.80
114.80

17.80
-8.20
5.80
-3.20
5.20
-5.80
-10.80

5.25
5.25
7.00
7.00

125
116
94
86

114.80
114.80
96.20
96.20

114.80
114.80
96.20
96.20

10.20
1.20
-2.20
-10.20

7.00
7.00
7.00
10.50

105
100
96
39

96.20
96.20
96.20
50.40

96.20
96.20
96.20
50.40

8.80
3.80
-0.20
-11.40

10.50
10.50
10.50
10.50

66
49
48
50

50.40
50.40
50.40
50.40

12.20
17.80
-8.20
5.80
-3.20
5.20
-5.80
10.80
10.20
1.20
-2.20
10.20
8.80
3.80
-0.20
11.40
15.60
-1.40
-2.40
-0.40

50.40
50.40
50.40
50.40

15.60
-1.40
-2.40
-0.40

Table 2. Results of the experiment

NOT Recovery Cyanide (mg/liter)

Air Pressure (Liter/min)

10.50

7.00

5.25

4.20
0

10

20

30

40

50

60

70

80

90

100

110

120

130

140

150

160

170

180

190

200

210

Not recovery Cyanide (mg/Liter)

Graphic 1. Not Recovery Cyanide

As we see in table 1, the less air pressure we give, the less recover cyanide we get. That means so far, our hypothesis is correct. Now we have to
prove it as we make the following analysis.

Initial Concentration
for CNAir/Solution
Proportion

Not Recovery
Cyanide

(Liter/minute)

(mg/liters)

4.20

5.25

7.00

10.50

172
202
176
190
181
120
109
104
125
116
94
86
105
100
96
39
66
49
48
50

250
Not recovered
Cyanide (%)
69%
81%
70%
76%
72%
48%
44%
42%
50%
46%
38%
34%
42%
40%
38%
16%
26%
20%
19%
20%

Table 3. Cyanide Percentage Recovered

Recovered
Cyanide (%)
31%
19%
30%
24%
28%
52%
56%
58%
50%
54%
62%
66%
58%
60%
62%
84%
74%
80%
81%
80%

Percentage of Recoved Cyanide (%)

Recovered Cyanide (%)

90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
4.20

5.25

7.00

10.50

Air Pressure (Liter/minute)

Graphic 2. Recovered Cyanide (%)

As we saw in the graphic above there is a direct linear relationship, because if we increase the air pressure
we obtain more cyanide from the solution, so the recovered percentage of the cyanide increase as well, just
as we expected.

ANOVA

As we can see in the graphic the interactions between this four different pressures are almost none. However at first
sight we can appreciate some interaction between the results of 5.25 pressure and 7 pressure Liter/minute
However if we go through the numbers in a more specific way, well see that the last number of 5.25 Liter/minute is
104 and the first number of 7 Liter/minute is 105, so there is really no interaction. The blue dots means where the mean
is placed and the grey ones are the variations. (All the results of the experiment).

In this graphic the little dots are the average of data, the horizontal line is the median and we can appreciate the
minimum and maximum value of every experiment, which give us the confidence intervals.
If there is no overlap in the data so that means every experiment is different form one another and there is no interaction
or interference with each other. (Significate difference)
As I said earlier, even though it seems there is an overlap, there is none.

If we see the histogram we know there is a tendency for a normal equation, and as we wanted almost all the data
concentrates in zero. However at the edges of the line tendency the model loses its strength. Our main goal would be
all the data concentrated in zero.

In this graphic we can see that the confidence intervals goes through 184.2 at 4.20 Liter/minute to 50.4 at 10.5
Liter/minute.

Again the difference between 5.25 Liter/minute and 7 Liter/minute pressure is almost imperceptive and that is why this
is the data closer that zero.

For the Dunnett I tested both 10.5 Liter/minute and 7 Liter/minute because I wanted to know if it would be some
significant difference and I also wanted to know if this particularly change could give more points with my boss in case
this would be an actual problem in the Company.
While the data is closer to zero, the less effect had the experiment with one another.

In this case, contrary of what we had been doing, if the data touches zero, that means there is a significant different in
the experiments.

In this six comparisons we can appreciate all the interactions the different air pressures have with one another. However
we can see that the 7.00 5.25 factor is closer to zero. That means that the relevance of the interaction in almost
inexistent. To the contrary of the pressure 10.50 4.20, where the relevance of different pressures is more relevance
to the Cyanide recovery.

One-way

ANOVA: Not Recovery Cyanide versus Air/Solution Proportion

Method
Null hypothesis.
All means are equal
Alternative hypothesis At least one mean is different
Significance level = 0.05
Equal variances were assumed for the analysis.
Factor Information
Factor
Levels
Air/Solution Proportion
4

Values
4.20, 5.25, 7.00, 10.50

Here we see four different levels that equals 4.20 Liter/minute, 5.25 Liter/minute, 7 Liter/minute and 10 Liter/minute.
Analysis of Variance
Source
Air/Solution Proportion
Error
Total

DF
3
16
19

Adj SS
46317
1442
47759

Adj MS
15439.1
90.1

F-Value
171.35

P-Value
0.000

As we see our F-Value is high which is good and our P-Value is zero which is perfection.

Model Summary
S
9.49210

R-sq
96.98%

R-sq (adj)
96.42%

R-sq(pred)
95.28%

Here we can see that the r-squared is actually awesome, because its more that 85% that is the minimum request.
And we have it almost 10% higher.
Means
Air/Solution
Proportion N
Mean
4.20
5
184.20
5.25
5
114.80
7.00
5
96.20
10.50
5
50.40
Pooled StDev = 9.49210

StDev
12.01
8.41
7.09
9.76

95% CI
(175.20, 193.20)
(105.80, 123.80)
(087.20, 105.20)
(041.40, 059.40)

Tukey Pairwise Comparisons

Grouping Information Using the Tukey Method and 95% Confidence
Air/Solution
Proportion
4.20
5.25
7.00
10.50

N
5
5
5
5

Mean
184.20
114.80
96.20
50.40

Grouping
A
B
C
D

Means that do not share a letter are significantly different.

Tukey Simultaneous Tests for Differences of Means
Difference,
Of Levels
5.25 - 4.20
7.00 - 4.20
10.50 - 4.20
7.00 - 5.25
10.50 - 5.25
10.50 - 7.00

Difference
of Means
-69.40
-88.00
-133.80
-18.60
-64.40
-45.80

SE of
Difference
6.00
6.00
6.00
6.00
6.00
6.00

(-086.59,
(-105.19,
(-150.99,
(-035.79,
(-081.59,
(-062.99,

95% CI
-52.21)
-70.81)
-116.61)
-01.41)
-47.21)
-28.61)

Individual confidence level = 98.87%

Again, our P-Value is almost zero every time.

Tukey Simultaneous 95% CIs

Fisher Pairwise Comparisons
Grouping Information Using the Fisher LSD Method and 95% Confidence
Air/Solution
Proportion
4.20
5.25
7.00
10.50

N
5
5
5
5

Mean
184.20
114.80
96.20
50.40

Grouping
A
B
C
D

Means that do not share a letter are significantly different.

T-Value
-11.56
-14.66
-22.29
-3.10
-10.73
-7.63

Adjusted
P-Value
0.000
0.000
0.000
0.032
0.000
0.000

Fisher Individual Tests for Differences of Means

Difference,
Of Levels
5.25 - 4.20
7.00 - 4.20
10.50 - 4.20
7.00 - 5.25
10.50 - 5.25
10.50 - 7.00

Difference
of Means
-69.40
-88.00
-133.80
-18.60
-64.40
-45.80

SE of
Difference
6.00
6.00
6.00
6.00
6.00
6.00

95% CI
(-082.13, -56.67)
(-100.73, -75.27)
(-146.53, -121.07)
(-031.33, -5.87)
(-077.13, -51.67)
(-58.53, -33.07)

T-Value
-11.56
-14.66
-22.29
-3.10
-10.73
-7.63

Adjusted
P-Value
0.000
0.000
0.000
0.007
0.000
0.000

Simultaneous confidence level = 81.11%

Fisher Individual 95% CIs

Dunnett Multiple Comparisons with a Control
Grouping Information Using the Dunnett Method and 95% Confidence
Air/Solution
Proportion
10.50 (control)
4.20
5.25
7.00

N
5
5
5
5

Mean
50.40
184.20
114.80
96.20

Grouping
A

Means not labeled with the letter A are significantly different from the control level mean.
Dunnett Simultaneous Tests for Level Mean - Control Mean
Difference,
Of Levels
4.20 - 10.50
5.25 - 10.50
7.00 - 10.50

Difference
of Means
133.80
64.40
45.80

Individual confidence level = 98.04%

SE of
Difference
6.00
6.00
6.00

95% CI
(118.24, 149.36)
(48.84, 79.96)
(30.24, 61.36)

T-Value
22.29
10.73
7.63

Adjusted
P-Value
0.000
0.000
0.000

Dunnett Simultaneous 95% CIs

Hsu Multiple Comparisons with the Best (MCB)
Hsu Simultaneous Tests for Level Mean - Smallest of Other Level Means
Difference,
Difference
SE of
Of Levels
of Means
Difference
95% CI
4.20 - 10.50 133.80
6.00
(00.00, 147.17)
5.25 - 10.50
64.40
6.00
(0.00, 77.77)
7.00 - 10.50
45.80
6.00
(0.00, 59.17)
10.50 - 7.00
-45.80
6.00
(-59.17, 0.00)

T-Value
22.29
10.73
7.63
-7.63

Adjusted
P-Value
0.000
0.000
0.000
0.000

Individual confidence level = 95.94%

Dunnett Multiple Comparisons with a Control

Grouping Information Using the Dunnett Method and 95% Confidence
Air/Solution
Proportion
7.00 (control)
4.20
5.25
10.50

N
5
5
5
5

Mean
96.20
184.20
114.80
50.40

Grouping
A

Means not labeled with the letter A are significantly different from the control level mean.
Dunnett Simultaneous Tests for Level Mean - Control Mean
Difference, Difference
Of Levels
of Means
4.20 - 7.00
88.00
5.25 - 7.00
18.60
10.50 - 7.00
-45.80

SE of
Difference
6.00
6.00
6.00

Individual confidence level = 98.04%

95% CI
(72.44, 103.56)
(03.04, 34.16)
(-61.36, -30.24)

T-Value
14.66
3.10
-7.63

Adjusted
P-Value
0.000
0.018
0.000

Conclusion:
In conclusion we can agree that our hypothesis is correct and as we said, if we increase the air pressure solution we
can recover more cyanide than if we use less air pressure solution.
In this experiment using 10.5 liters/min. give us the best combination to accomplish our goal.

Referencies:
Congreso Internacional de Metalurgia (Caracas)
Metalurgia 1997 (Acapulco)
Tesis de Posgrado ITS 1997: Concurso Nacional de Creatividad Posgrado 1997
CYANIDE RECOVERY:
http://www.sgs.mx/~/media/Global/Documents/Flyers%20and%20Leaflets/SGS-MIN-WA016-Cyanide-RecoveryComparison-EN-11.pdf