SIMPLIFIED  “FLOTATION  DE-­‐INKING”  OF  NEWSPAPER

NAUMAN  MITHANI  (CHEM  317  -­‐  2010  SPRING)

1

• de-­‐ink  newspaper  as  per  the  technique  of  flota%on  de-­‐inking

• scaled  down  &  adapted  for  lab  environment                                  improvised  equipment

• focus  on  op#misa#on  of  variables:

✓ duraAon  of  flotaAon

✓ concentraAon  of  flotaAon  agent

-­‐ temperature

-­‐ consistency  (cs.  %)

PURPOSE

if  possible

2

• paper,  a  ubiquitous  material

• a  disposable  material

• 33%  to  50%  of  solid  municipal  waste*

➡ a  crucial  parameter  in  the  long-­‐term  environmental  sustainability  of  a  large  populaXon  centre

✤ so  it  must  be  recycled,  which  in  turn  consumes...

✤ water

✤ electricity

✤ &  other  (environmental)  resources

✤ sustainability  &  expansion  if:

✤ high  performance  &  high  quality  of  recycled  paper

BACKGROUND:    paper  facts

0

25

50

75

100

USA (2004) Canada (2006)

4.5

50

13.5

50

pape

r (m

illion

ton

nes)

un-recoveredrecovered

* Nie et al. Environmental Engineering and Policy (1998) vol. 1 (1) pp. 47-58 3

• Paper,  essenXally,  is  a  felted  sheet  of  cellulose  fibres

• actually,  it  is  a  complex  mixture  of    chemical  addiXves,  fillers,  bonding  agents  (upto  ~40  %  weight)

• Cellulose:  (C6H10O5)n

• bears  a  surfeit  of  O-­‐  and  OH  groups                  network  of  H-­‐bonds

• highly  hydrophilic

• zero  water-­‐contact  angle

INTRO:  chemical  composiXon  of  paper

hydrophilic and hydrophobic surfaces. The contact angle

is measured from the surface through the liquid to the

tangent of the liquid surface as indicated by the arrows. Hydrophilic Hydrophobic

4

• Ink  applies  to  paper...

• ...  by  absorpXon  which  then  dries/sets  in  the  fibre

• ...  OR...  by  fusing  which  then  cools  and  bonds  to  the  fibre

• different  composiXon  for  different  printed  materials:

• NEWSPRINT  INK:  mineral/vegetable  oil  (45  -­‐  60%)  and  resin  (5  to  35%)

• the  oil  serves  as  the  vehicle  for  the  pigment

• black:  inorganic  carbon  (graphite)  blacks

• colours:  organic  pigments

• immiscible  oil  +  (co)polymer  resin  +  carbon  black                          ink  is  H-­‐phobic

• water  contact  angle  is  >80°.

INTRO:  the  ink  (toner)

hydrophilic and hydrophobic surfaces. The contact angle

is measured from the surface through the liquid to the

tangent of the liquid surface as indicated by the arrows. Hydrophilic Hydrophobic 5

• (froth)  flotaXon  de-­‐inking  is  the  more  popular  de-­‐inking  method

• bejer  than  wash  de-­‐inking  (laundry-­‐esque)  since  it  can  also  remove  larger  parXcles:  10  -­‐  400  μm  (diameter)  in  addiXon  to  fine  ink  &  filler  parXcles.

• overall:  lesser  quanXXes  of  water  needed

• overall:  higher  quality  of  recycled  paper

• schemaXc  below  shows  industrial  de-­‐inking  plant  setup

INTRO:  flotaXon  de-­‐inking

source: Wikipedia6

• first  step  -­‐  pulping:  the  paper  is  disintegrated  in  water  to  form  a  pulp  slurry

• collector  (foaming  agent  +  surfactant)  is  added  to  pulp  slurry

• air  (bubbles)  is  introduced

• the  H-­‐phobic  ink  binds  to  collector  +  air  bubble

• carried  to  the  surface  -­‐  REJECT  stream

• forms  “contaminant”-­‐laden  foam/froth

• which  is  then  removed  or  scraped-­‐off

• the  flotaXon  is  only  1  in  a  series,  there  are  2  series

INTRO:  the  flotaAon  de-­‐inking  process

7source: Wikipedia

• actuality:  conXnuous  FEED  of  pulp  slurry  +  collector  mixture  into  such  a  chamber  where  air  is  con#nuously  supplied

• only  1  improvised  flota#on  cell  shall  be  used  here

8source: Deng et al. “SURFACTANT SPRAY”...Georgia Institute of Technology (2004)

schemaXc  of  industrial  flotaXon  cell  setup

The second parameter, cleanliness efficiency, is calculated as follows:

eq. 2: Cleanliness efficiency =FEED count ! ACCEPT count( )

FEED count" 100

It has been found that the operational parameters of a flotation de-inking process must balance between high

fibre yield and high cleanliness efficiency: if a high fraction of the FEED is subjected to rejection then the remainder of

the (processed) sample may be very clean; naturally, the down side is low fibre yield; and vice versa.

• Experimental overview

This series of experiments seeks to build upon the paper on lab-scale flotation de-inking by Venditti, R. [J.

Chem. Ed. 81(5), 693]: “A Simple Flotation De-Inking Experiment for the Recycling of Paper” [9]. The experimental

overview is cited as follows:

“This paper describes a laboratory exercise for the f lotation de-inking of wastepaper. The exercise consists of disintegrating

printed wastepaper in a blender and then removing the ink or toner contaminants by pumping air bubbles through the

suspension using an aquarium pump or other source of air bubbles. Foam is taken off the top of the container that is rich in

ink (the reject sample) while the cleaned fiber remains in the container (the accept sample). After the experiment the accept and

reject samples are analyzed for ink concentration and for fiber content.

Common, inexpensive equipment and no chemicals (other than a surfactant to enhance foaming) are needed for the

exercise.” [9]

Figure 3: A schematic drawing of the laboratory flotation

experiment.

6

EWFTiBook:Users:fedosky:Desktop:_Online:JCE May 2004:JCE0504_Web Supplements:JCE2004p0693W.doc

2

A Simple Flotation De-inking Experiment for the Recycling of Paper

Richard A. Venditti

Associate ProfessorNorth Carolina State UniversityDepartment of Wood and Paper ScienceRaleigh NC, 27695-8005Telephone: (919) 515-6185Fax: (919) 515-6302Email: richard_venditti@ncsu.edu

Lab Summary

Flotation de-inking is used in paper recycling processes to preferentially remove hydrophobic contaminants

such as inks and toners from a slurry of fibers in an aqueous phase. In the process, fine air bubbles are

introduced into the suspension and the hydrophobic contaminants preferentially attach to the bubble-water

interfaces and float to the surface. The foam on the top of the surface laden with contaminant is skimmed

away resulting in the separation.

This paper describes a laboratory exercise for the flotation de-inking of wastepaper. The exercise consists

of disintegrating printed wastepaper in a blender and then removing the ink or toner contaminants by

pumping air bubbles through the suspension using an aquarium pump (Figure 1) or other source of air

bubbles. Foam is taken off the top of the container that is rich in ink (the reject sample) while the cleaned

fiber remains in the container (the accept sample). After the experiment the accept and reject samples are

analyzed for ink concentration and for fiber content.

Tubing Foam laden with ink

Air Pump Dispersed Wastepaper

Large Plastic Tray

Figure 1. Schematic Drawing of the Laboratory Flotation Experiment.

Common, inexpensive equipment and no chemicals (other than a surfactant to enhance foaming) are needed

for the exercise. The experiment is useful for middle/high school science courses or introductory level

college environmental, chemical engineering, or chemistry courses in need of a simple experiment that

schemaXc  of  the  improvised  flotaXon  cell  

(

photo  of  an  industrial  flotaXon  cell  using  “surface  surfactant  spray”photo  of  the  improvised  flotaXon  cell

• fibre  yield,  Y  (%):  

• based  on  the  law  of  mass  conservaAon:  (F)EED    =    (A)CCEPT  +  (R)EJECT

• fibre  yield  based  on  (F)EED  and  (A)CCEPT  stream:    

• fibre  yield  based  on  (F)EED  and  (R)EJECT  stream:  

• fibre  yield  based  on  (A)CCEPT  and  (R)EJECT  stream:

9

INTRO:  flotaAon  de-­‐inking  performancedry mass of accepted solids

dry mass of total solids fed into process× 100

YFA %( ) = AF× 100

YFR %( ) = F − RF

× 100

YAR %( ) = AA + R

× 100

• contaminant  removal  OR  cleanliness  efficiency:

• based  on  the  contaminant  or  ink  spec  count  of  the  (F)EED  and  (A)CCEPT  streams:

cleanliness efficiency = FEED count - ACCEPT countFEED count

× 100

• household  blender

• generic  1  litre  aqua#c  tank  pump  coupled  with  an  air  diffuser

• Paper:  same  pages  from  mulXple  copies  of  a  standard  newspaper  (SFU  Peak)  printed  on  52/75  e-­‐brite  paper  (NORPAC,  Burnaby,  BC)

• Ink:  standard  newspaper  ink  (Sun  Chemical,  Richmond,  BC)  -­‐  exact  formulaXon  could  not  be  known

• foaming  agent  (collector):  “BRD2345”,  a  proprietary  foaming  agent  (Buckman  Laboratories,  Memphis,  USA)

• microscope  &  soQware  for  ink  spec  size  and  density  characterisaXon:  

• MoAc  B-­‐5  Professional  Series  with  MoAc  images  Advanced  3.0  by  Micro-­‐OpXc  Industrial  group  (Richmond,  BC)

10

EXPERIMENTAL:  notable  MATERIALS  (common,  inexpensive)

1. cut  2.25g  worth  of  paper  into  3  ×  3  cm  squares  and  place  them  in  blender  with  400  mL  of  tap  water  (pH:  6.51)  of  desired  temperature

2. blended  for  3  min.

3. added  desired  amount  of  foaming  agent  to  the  pulp  slurry  and  sXrred  gently  at  length  (>  5  min.)

4. divided  the  pulp  slurry  into  two  beakers  labelled  FEED  stream  &  ACCEPT-­‐to-­‐be  stream

5. filled  the  ACCEPT-­‐to-­‐be  stream  beaker  to  desired  volume  (maintained  temperature)  to  control  consistency  (cs.  %)

6. placed  the  ACCEPT-­‐to-­‐be  stream  into  a  relaXvely  large  plasXc  tray

7. the  air  diffuser  connected  to  the  air  pump  (by  tubing)  was  placed  into  the  beaker

8. the  pump  was  turned  on  to  mark  the  start  of  the  flotaXon

11

EXPERIMENTAL:  METHOD  (overview)

9. the  generated  foam  was  periodically  scraped  off  (REJECT  stream)

10. arer  a  certain,  desired  Xme  the  pump  was  turned  off  to  mark  the  end  of  the  flotaXon

11. the  contents  of  the  plasXc  tray  (REJECT  stream),  the  remnants  in  the  ACCEPT-­‐to-­‐be  beaker  (now  the  ACCEPT  stream)  &  the  contents  of  the  FEED  stream  were  filtered  (Whatman  1:  110  mm)

12. the  contents  of  each  stream  were  dried  overnight  at  105  °C

13. the  mass  of  the  contents  (fibres)  of  each  stream  was  measured

14. contaminant  (ink  spec)  characterisaXon  was  ajempted  using  the  microscope  setup

12

EXPERIMENTAL:  METHOD  (overview)(contd.)

The second parameter, cleanliness efficiency, is calculated as follows:

eq. 2: Cleanliness efficiency =FEED count ! ACCEPT count( )

FEED count" 100

It has been found that the operational parameters of a flotation de-inking process must balance between high

fibre yield and high cleanliness efficiency: if a high fraction of the FEED is subjected to rejection then the remainder of

the (processed) sample may be very clean; naturally, the down side is low fibre yield; and vice versa.

• Experimental overview

This series of experiments seeks to build upon the paper on lab-scale flotation de-inking by Venditti, R. [J.

Chem. Ed. 81(5), 693]: “A Simple Flotation De-Inking Experiment for the Recycling of Paper” [9]. The experimental

overview is cited as follows:

“This paper describes a laboratory exercise for the f lotation de-inking of wastepaper. The exercise consists of disintegrating

printed wastepaper in a blender and then removing the ink or toner contaminants by pumping air bubbles through the

suspension using an aquarium pump or other source of air bubbles. Foam is taken off the top of the container that is rich in

ink (the reject sample) while the cleaned fiber remains in the container (the accept sample). After the experiment the accept and

reject samples are analyzed for ink concentration and for fiber content.

Common, inexpensive equipment and no chemicals (other than a surfactant to enhance foaming) are needed for the

exercise.” [9]

Figure 3: A schematic drawing of the laboratory flotation

experiment.

6

EWFTiBook:Users:fedosky:Desktop:_Online:JCE May 2004:JCE0504_Web Supplements:JCE2004p0693W.doc

2

A Simple Flotation De-inking Experiment for the Recycling of Paper

Richard A. Venditti

Associate ProfessorNorth Carolina State UniversityDepartment of Wood and Paper ScienceRaleigh NC, 27695-8005Telephone: (919) 515-6185Fax: (919) 515-6302Email: richard_venditti@ncsu.edu

Lab Summary

Flotation de-inking is used in paper recycling processes to preferentially remove hydrophobic contaminants

such as inks and toners from a slurry of fibers in an aqueous phase. In the process, fine air bubbles are

introduced into the suspension and the hydrophobic contaminants preferentially attach to the bubble-water

interfaces and float to the surface. The foam on the top of the surface laden with contaminant is skimmed

away resulting in the separation.

This paper describes a laboratory exercise for the flotation de-inking of wastepaper. The exercise consists

of disintegrating printed wastepaper in a blender and then removing the ink or toner contaminants by

pumping air bubbles through the suspension using an aquarium pump (Figure 1) or other source of air

bubbles. Foam is taken off the top of the container that is rich in ink (the reject sample) while the cleaned

fiber remains in the container (the accept sample). After the experiment the accept and reject samples are

analyzed for ink concentration and for fiber content.

Tubing Foam laden with ink

Air Pump Dispersed Wastepaper

Large Plastic Tray

Figure 1. Schematic Drawing of the Laboratory Flotation Experiment.

Common, inexpensive equipment and no chemicals (other than a surfactant to enhance foaming) are needed

for the exercise. The experiment is useful for middle/high school science courses or introductory level

college environmental, chemical engineering, or chemistry courses in need of a simple experiment that

• a  schemaXc  of  the  experimental  setup  is  shown  here

• the  ink  specs  were  too  small  &  numerous  for  the  naked  eye

• could  not  carried  out  with  the  microscope:  two  sample  photos  from  different  experiments  are  shown  below

• even  at  10X,  the  specs  are  too  small  and  numerous  for  a  manual  count  -­‐  the  sorware  did  not  have  the  ability  to  conduct  a  parXcle/spec  count

• at  >10X,  there  was  insufficient  light  for  any  viewing 13

RESULTS  &  DISCUSSION:   INK  SPEC  COUNT  FAILURE

• consistency,  cs.  %  is  defined  as:                                                                                                                                              where  ρ(water)  =  1  g/mL

14

RESULTS  &  DISCUSSION:   effect  of  CS.  %

cs. %( )= mass of solids (grams)volume of water (mL)

× 100

T (°C) cs. (%)vol. (mL)

conc. (mL/g)

time (min.)

avg. Y (%)

σ (%)

23 0.296 5.00 2.22 10.023 0.468 5.00 2.22 10.0

48.9 4.728.5 19.2

• as  the  cs.  %  is  raised  by  58%,  the  yield  falls  by  the  same  number:  ΔY(%)  /  Δcs.(%)  =  1

• expectedly,  since  there  is  a  greater  amount  of  fibre  per  unit  volume  of  water...

• ...  there  is  more  that  may  be  carried  off  into  the  REJECT  stream.

• results  are  inconclusive  since  there  are  insufficient  data  points.

0

10

20

30

40

50

60

0.20 0.26 0.32 0.38 0.44 0.50

average Y (%) vs. cs. (%)

avg

. Y (%

)

cs. (%)

• as  the  temperature  is  raised  by  52%,  the  fibre  yield  falls  by  26%:                                                                      ΔY(%)  /  ΔT  (°C)  =  0.5

• expectedly,  as  there  is  greater  convecXon  and  greater  thermal  moXon...

• ...  there  is  a  greater  amount  that  goes  up  &  out  as  REJECT  stream

• results  are  inconclusive  since  there  are  insufficient  data  points

15

RESULTS  &  DISCUSSION:   effect  of  TEMPERATURE

T (°C) cs. (%)vol. (mL)

conc. (mL/g)

time (min.)

avg. Y (%)

σ (%)

23 0.468 5.00 2.22 10.0

35 0.468 5.00 2.22 10.0

28.5 0.6

21.2 20.5

0

10

20

30

40

50

20 25 30 35 40

average Y (%) vs. temperature (°C)

avg

. Y (%

)

temperature (°C)

16

RESULTS  &  DISCUSSION:   effect  of  TIME  (min.)

T (°C) cs. (%)vol. (mL)

conc. (mL/g)

time (min.)

avg. Y (%)

σ (%)

23 0.468 5.00 2.22 10.023 0.468 5.00 2.22 5.023 0.468 5.00 2.22 3.023 0.468 5.00 2.22 2.023 0.468 5.00 2.22 1.523 0.468 5.00 2.22 1.0

28.5 19.2

35.8 18.4

40.9 31.1

51.6 20.6

54.4 15.1

64.2 0.2

0

10

20

30

40

50

60

70

80

0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0

R² = 0.99

average Y (%) vs. time (min.)

avg

. Y (%

)

time (min.)

• downward  trend  is  expected

• trend  is  exponenXal:  as  Xme  goes  by...

• ...  fibre  +  water  are  REJECTED

• the  layer  of  foam  becomes  thicker  &  thicker  -­‐  reaches  deeper  into  the  beaker

• acts  as  physical  filter/barrier  for  remaining  fibres...

• ...  slows  further  fibre  REJECTION

• as  per  the  trend,  ideal  flotaXon  duraXon:  0  <  t  (min.)  ≪  1  -­‐  unrealisXc

• table  shows  a  significant  increase  below  3  min.  -­‐  the  1.5  -­‐  2.0  min.  region

17

RESULTS  &  DISCUSSION:  effect  of  CONCENTRATION  of  FOAMING  AGENT

T (°C) cs. (%)vol. (mL)

conc. (mL/g)

time (min.)

avg. Y (%)

σ (%)

23 0.468 5.00 2.22 1.5

23 0.468 4.00 1.78 1.5

23 0.468 3.00 1.33 1.5

23 0.468 2.00 0.89 1.5

23 0.468 1.00 0.44 1.5

23 0.468 0.50 0.22 1.5

23 0.468 0.25 0.11 1.5

54.7 15.3

58.8 0.2

58.2 0.8

48.9 4.7

58.3 5.8

53.8 41.5

57.2 40.9

0

10

20

30

40

50

60

70

80

90

100

0 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00

R² = 0.0001

Y (%) vs. vol. of BRD2345 foaming agent

avg.

Y (%

)

volume of BRD2345 (mL)

• fibre  yield  is  not  affected  by  concentraXon  of  foaming  agent

• a  volume  as  low  as  0.25  mL  in  240  mL  of  water  may  be  sufficient  for  flotaXon

• 1:1000  of  foaming  agent:water

• lower  cost  of  operaXon

18

FURTHER  DISCUSSION• individual  high  error  margins:  fibre  loss

• wet  fibre  sXcks  to  every  container/vessel  -­‐  difficult  to  wash  down  -­‐  scrubbing  down  is  not  possible  during  an  experiment

• ∴  fibre  loss  in  every  transfer/step  e.g.  division  into  separate  FEED  and  ACCEPT  streams

• compounded  with  each  subsequent  step  -­‐  the  further  a  step  is  from  the  FEED  stream,  the  higher  the  loss  (reason  for  the  significantly  different  YFR)

• suggesAon:  use  a  flotaXon  vessel  with  rough  interior  walls  or  Teflon

• other  a^empts  at  ink  spec  count:  failure

• pulped  and  intact  squares  were  placed  in  various  solvents  to  ajempt  a  leeching-­‐out  for  subsequent  UV-­‐Vis  spectroscopic  analysis

• ...  benzene,  toluene,  xylenes,  DMSO,  acetonitrile,  polypropylene  glycol,  glycerol,  methanol  &  ethanol...

• graphite  is  virtually  insoluble  -­‐  all  ajempts  failed

• suggesAon:  dedicated  system  e.g.  flatbed  scanner/CCD  camera  +  image  analysis  sorware  e.g.  Apogee  Spec*Scan  2000

19

CONCLUSION• cs.  %:  higher  fibre  loss  with  higher  cs.  %

• temperature:  higher  fibre  loss  with  higher  temperature

• dura#on:  a  realisXc  opXmum  of  1.5  min.  was  determined  with  a  fibre  yield  of  54.4%

• concentra#on  of  foaming  agent  BRD2345:  has  no  effect  on  fibre  yield

• suitable  concentraXon:  1:1000  in  water

• results  are  only  par#ally  valid:  only  one  of  two  performance  criteria  could  be  obtained

• effect  of  experimental  variables  on  contaminant  count  could  not  be  ascertained

• dedicated,  proper  equipment  is  needed

20

ACKNOWLEDGEMENTS

• Dr.  Richard  Vendie  of  the  Wood  &  Paper  Science  Dept.  of  North  Carolina  State  University  for  his  guidance  and  provision  of  the  foaming  agent  BRD2345.

• Dr.  Dev  Sharma,  for  his  general  assistance

• SFU  Dept.  of  Chemistry  for  their  generous  funding  and  making  this  only  a  2-­‐credit  course

• Tahir,  for  his  ears

• Jasbir,  for  his  complete  experimental  failure  thereby  reassuring  my  that  parXal  failure  is  not  so  bad

• Yuen,  for  never  being  here  -­‐  the  extra  benchspace  was  invaluable

• Neil  Draper,  for  reminding  me  that  pH  of  water  slowly  decreases  in  open  air

• ...  Dr.  Goyan,  for  the  ?-­‐grade...

21

REFERENCES• [1]   Breedlove.  The  analysis  of  ball-­‐point  inks  for  forensic  purposes.  Journal  of  chemical  educa#on  (1989)  vol.  66    (2)  pp.  170

• [2]   McCullagh  and  Ramos.  Separa#on  of  the  Carotenoid  Bixin  from  Anna^o  Seeds  Using  Thin-­‐Layer  and  Column    Chromatography.  Journal  of  chemical  educa#on  (2008)  vol.  85  (7)  pp.  948

• [3]   Quach  et  al.  An  Improved  Method  for  the  Extrac#on  and  Thin-­‐Layer  Chromatography  of  Chlorophyll  A  and  B    from  Spinach.  Journal  of  chemical  educa#on  (2004)  vol.  81  (3)  pp.  3

• [4]   Harmon  et  al.  Crime  Scene  Inves#ga#on  in  the  Art  World:  The  Case  of  the  Missing  Masterpiece.  Journal  of  chemical  educa#on  (2009)  vol.  86  (7)  pp.  817

• [5]   Judd.  News  from  Online:  Chemistry  and  Art.  Journal  of  chemical  educa#on  (2001)  vol.  78  pp.  1322

• [6]   Miller  et  al.  Recycling  office  waste  -­‐  Recovered  paper  deinking  research  at  the  Univeristy  of  Utah.  Paper  Recycling  Challenge  -­‐  Deinking  &  Bleaching  (1997)

• [7]   Deng  and  Zhu.  SURFACTANT  SPRAY:  A  NOVEL  TECHNOLOGY  TO  IMPROVE  FLOTATION  DEINKING  PERFORMANCE.  Internal  Technical  Report  of  School  of  Chemical  &  Biomolecular  Engineering  Georgia  Ins#tute  of  Technology    (2004)

• [8]   Jeffries  et  al.  COMPARISON  OF  ENZYME  ENHANCED  WITH  CONVENTIONAL  DEINKING  OF  XEROGRAPHIC  AND  LASER  PRINTED  PAPER.  TAPPI  Journal  (1994)  vol.  77  (4)  pp.  173-­‐179

• [9]   Vendie.  A  Simple  Flota#on  De-­‐Inking  Experiment  for  the  Recycling  of  Paper.  Journal  of  chemical  educa#on  (2004)  vol.  81  (5)  pp.  693-­‐693

• [10]   Quigley  and  Qi.  A  chemistry  whodunit:  Forensic  examina#on  of  pen  inks.  Journal  of  chemical  educa#on  (1991)  vol.  68  (7)  pp.  597

• [11]   Beneven#  et  al.  Influence  of  surfactant  concentra#on  on  the  ink  removal  selec#vity  in  a  laboratory  flota#on    column.  Interna#onal  Journal  of  Mineral  Processing  (2008)  vol.  87    (3-­‐4)  pp.  134-­‐140

• [12]   Ifa  et  al.  Forensic  analysis  of  inks  by  imaging  desorp#on  electrospray  ioniza#on  (DESI)  mass  spectrometry.  The  Analyst  (2007)  vol.  132  (5)  pp.  461-­‐467

• [13]   Pan  and  Nguyen.  Development  of  the  Photoacous#c  Rapid-­‐scan  FT-­‐IR-­‐based  method  for    measurement  of  ink  concentra#on  on  printed  paper.  Analy#cal  chemistry  (2007)  vol.  79  pp.  2259-­‐2265

• [14]   Liu  et  al.  Inves#ga#on  on  solubility  of  polymeric  binder  of  xerographic  toner  and  de-­‐inking  by  emulsion  process.  Journal  of  the  Taiwan  Ins#tute  of  Chemical  Engineers  (2009)  vol.  40  (1)  pp.  84-­‐90

• [15]   Ware.  Prussian  Blue:  Ar#sts'  Pigment  and  Chemists'  Sponge.  Journal  of  chemical  educa#on  (2008)  vol.  85  pp.  612–621

• [16]   Puddington  et  al.  Ink  removal  from  waste  paper.  United  States  Patent  no.  4,076,578    (1978)

• [17]   Nie  et  al.  The  effect  of  ink  types  and  prin#ng  processes  on  flota#on  deinking  efficiency  of  wastepaper  recycling.  Environmental  Engineering  and  Policy  (1998)  vol.  1  (1)  pp.  47-­‐58

• [18]   Bohonowych  et  al.  Newspapers  and  Newspaper  Ink  Contain  Agonists  for  the  Ah  Receptor.  Toxicological  sciences  (2008)  vol.  102  (2)  pp.  278-­‐290

• [19]   Nishiyama  et  al.  Crystal  Structure  and  Hydrogen-­‐Bonding  System  in  Cellulose  I  [beta]  from  Synchrotron  X-­‐ray  and  Neutron  Fiber  Diffrac#on.  Journal  of  the  American  Chemical  Society  (2002)  vol.  124  (31)  pp.  9074-­‐9082

• [20]   Bourdages.  Paper  Recycling  in  Canada  -­‐  A  New  Reality.  Government  of  Canada  publica#ons  (1993)  pp.  1-­‐10

• [21]   Paper  Recycling  Associa#on.  Overview  of  the  recycling  industry.  Annual  publica#ons    (2007)  pp.  1-­‐4

• [22]   huber-­‐group.  Newspaper  inks  and  the  environment.  Huber  Group  Technical  informa#on  (2003)  pp.  1-­‐4

• [23]   Fairbank  et  al.  Effects  of  recovered  paper  quality  and  deinking  process  parameters  on  dirt  levels  in  newsprint.    Pulp  &  Paper  Canada  (2006)  vol.  107  pp.  12