An Efficient, Selective Process For The Conversion of Glycerol To Propylene Glycol Using Fixed Bed Raney Copper Catalysts

Subscriber access provided by Lulea University of Technology
Full Paper
An Efficient, Selective Process for the Conversion of Glycerol
to Propylene Glycol Using Fixed Bed Raney® Copper Catalysts
Setrak K Tanielyan, Norman Marin, Gabriela Alvez, Ramesh Bhagat,
Balaraju Miryala, Robert L. Augustine, and Stephen Schmidt
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/op400123f • Publication Date (Web): 06 Aug 2013
Downloaded from http://pubs.acs.org on August 18, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Organic Process Research & Development is published by the American Chemical

Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 25 Organic Process Research & Development
1
2
3
4
5
6
7
8
An Efficient, Selective Process for the Conversion
9
10
11
12 of Glycerol to Propylene Glycol Using Fixed Bed
13
14
15
16
17
Raney  Copper Catalysts
18
19
20
21 Setrak K. Tanielyan,a Norman Marin,a Gabriela Alvez,a Ramesh Bhagat,a Balaraju Miryala,a
22
23
24 Robert L. Augustinea* and Stephen R. Schmidtb
25
26
27 a
Center for Applied Catalysis, Department of Chemistry and Biochemistry, Seton Hall
28
29
30
University, South Orange, NJ 07079, USA. bW.R. Grace and Co.-Conn. 7500 Grace Drive,
31
32 Columbia MD 21044, USA
33
34
35 Raney  is a trademark, registered in the United States and/or other countries, of W.R. Grace &
36
37
38 Co.-Conn.
39
40
41 Keywords: Glycerol, Propylene Glycol, 1,2-Propanediol, Raney Copper, Hydrogenolysis,
42
43
44 Fixed Bed Reaction
45
46
47
48
49
50
51
52
53
54
55
56
57
58 1
59
60
ACS Paragon Plus Environment
Organic Process Research & Development Page 2 of 25
1
2
3
ABSTRACT: Propylene glycol is formed in yields of up to 95% at 100% conversion using a
4
5 Raney  Cu catalyst in a fixed bed reactor. The reaction uses an 80% aqueous glycerol solution
6
7 and a hydrogen pressure of 600 psi. The primary by-product is ethylene glycol formed in 1-3%
8
9 yield. In a reaction run continuously for 24 d using a sample of a commercial preparation of
10
11 Raney Cu, the selectivity to propylene glycol at 100% glycerol conversion was 94.6% with a
12
13 Space Time Yield (STY) of 0.49 g of 1,2-propylene glycol/cc Raney Cu/h. Ethylene glycol
14
was formed in 2.5% yield while methanol, ethanol, n-propanol 1,3-propylene glycol and acetol
15
16 were present in less than 1% yield.
17
18
19
20
21
22 TOC Graphic
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 2
59
60
1
2
3
1. INTRODUCTION
4
5
6
7 The increasing world-wide production of biodiesel by the transesterification of fats and oils has
8
9 resulted in a corresponding increase in the production of glycerol. One gallon of glycerol is
10
11
12 produced for every nine gallons of biodiesel. Even though there are over 2000 different
13
14 applications for which glycerol is used,1 new high volume commercial processes are needed to
15
16 improve the economics of biodiesel production. Some early work in this area concentrated on
17
18
19 the selective oxidation of glycerol2 to compounds such as dihydroxyacetone, hydroxypyruvic
20
21 acid and glyceric acid. For a variety of reasons this approach was not sufficiently effective to
22
23
utilize much of the glycerol which was being produced. Other research was directed toward
24
25
26 finding commercially applicable procedures for the conversion of glycerol into commodity
27
28 chemicals.3-5 Processes have been developed which use a glycerol feedstock to replace the
29
30
31
petrochemical feedstock for the preparation of acrolein6 and epichlorohydrin.7 More recently
32
33 procedures have been reported for converting glycerol to glycerol carbonate,8 a practical,
34
35 ecofriendly solvent, and glycerol alkyl ethers9 which are safe and effective fuel additives.
36
37
38
39 The most recent research efforts, though, have been directed to developing practical methods for
40
41 the conversion of glycerol (GLY) to 1,2-propylene glycol (PG) which is used primarily in
42
43 polymer synthesis but is also an ingredient in antifreeze and de-icing solutions. Since PG is
44
45
46 considered as Generally Recognized As Safe (GRAS) by the USFDA, it has also found use in
47
48 pharmaceuticals, personal care products and as a preservative in food.
49
50
51
52
The commonly proposed mechanistic pathways for the conversion of glycerol to glycols are
53
54 shown in Scheme 1. The formation of PG apparently takes place through an initial dehydration
55
56 of one of the primary hydroxyl groups on GLY to give the enol, 1, which is in equilibrium with
57
58 3
59
60
1
2
3
Scheme 1. Glycerol hydrogenolysis reactions
4
5
6 OH O OH
7 H2
8 C C CH
9 H2C 1 CH2 OH H3 C CH2 OH H3C CH2 OH
ACT PG
10
11 -H2O
12
13 OH
14
CH -H2O CH CH2
15 O
HO CH2 CH2 OH HO CH CH2 OH CH CH2 OH
16 2 3
GLY
17 -H2O H2
18 -H2
19 H2
20 OH CH CH2
21 O CH CH2 HO CH2 CH2 OH
H ACR
22 CH 1,3PG
O CH CH2 O
23 4
24
25 -CO
OH
26
H B
27 CH
28 O CH CH2 O
29 4
30 OH OH
31
H2C O H2
32 + CH CH2
33 HO CH H3C OH + HO CH2
34 5 EG
35 -H2O
36
O
37 CH2
H2
38 HOCH2CH3 C CH3 HO CH
39 H 6
40
41
42 the hydroxy ketone, acetol, (ACT). Hydrogenation of ACT gives PG. Dehydration of the
43
44 secondary hydroxyl group in GLY gives the enol, 2, which is in equilibrium with the hydroxyl
45
46
47 aldehyde, 3. Hydrogenation of 3 produces 1,3-propylene glycol (1,3PG) while dehydration
48
49 gives acrolein (ACR). One of the most common by-products in this reaction is ethylene glycol
50
51 (EG). One possible route to its formation is that initiated by the dehydrogenation of GLY to
52
53
54 glyceraldehde (4). A retro-Aldol reaction leads to the formation of formaldehyde and the enol, 5,
55
56 which on hydrogenation give EG and MeOH. Other potential routes to EG are the direct
57
58 4
59
60
1
2
3
hydrogenolysis of a C-C bond of GLY and the decarbonylation of 4 which would form EG
4
5
6 directly. A dehydration-hydrogenation sequence on EG leads to the formation of EtOH.
7
8
9 The catalysts involved in the conversion of GLY to PG generally contain Ru, Ni or Cu.
10
11
12 Supported Ru was used in conjunction with an acidic ion exchange resin to facilitate the initial
13
14 dehydration step.10-12 Other reports cited the use of supported Ru catalysts which also contained
15
16 Pt13 Au,13 or Fe.14 Supported Ni catalysts included Ni/C,15 Ni-Ce/C,16 and Ni/NaX.17 Co/MgO
17
18
19 was also used as a catalyst for this reaction.18 The most commonly used catalysts, however,
20
21 contained copper,19 usually either supported on ZnO,20-24modified with Cr25-29or as a component
22
23 in a mixed oxide system30-32 such as copper-chromite (Adkins Catalyst) (Cu-CrO).33-38 Most of
24
25
26 these reports described the use of batch processes for this reaction but some, primarily patents,
27
28 also discussed the use of continuous reactions.19,20,31,34,38
29
30
31
32
It is difficult to compare these procedures since they were run using a wide range of reaction
33
34 conditions: glycerol concentrations ranging from 10% aqueous GLY to pure GLY and technical
35
36 grade GLY; reaction temperatures from 140 °C to 270 °C; hydrogen pressures from 150 psi to
37
38
39 1500 psi with one patent31 describing the use of 3700-4400 psi of H2. The batch reactions were
40
41 run for 5 to 40 hours giving widely different results, reporting GLY conversions ranging from
42
43 23% to 100% and selectivities to PG from 20% to 95% as well as giving varying amounts of by-
44
45
46 products, primarily EG, ACT, n-propyl alcohol, ethanol and methanol. The continuous
47
48 processes utilized primarily copper containing catalysts, again with differing reaction conditions
49
50
making any comparisons difficult.
51
52
53
54 Probably the most extensive study of batch and continuous reactions for this transformation was
55
56 that reported by Suppes33-40 which used Cu-CrO as the catalyst. His approach was unique in that
57
58 5
59
60
1
2
3
it involved a two-stage reaction sequence in which ACT was formed first and then hydrogenated
4
5
6 in a separate step. Both batch39 and continuous reactions40 were used in this approach.
7
8
9 Little has been published on the use of Raney catalysts for the conversion of GLY to PG. One
10
11
12 paper described using Raney Ni as the catalyst for batch reactions comparing the extent of PG
13
14 formation between pure GLY and the GLY obtained from a biodiesel reaction. 41 Another
15
16
17 reported using Raney Ni in a batch process to yield selectivities of 77% PG, 15% ethanol and
18
19 8% CO2 at 63% conversion after 20 h. 42 In a survey of catalysts screened for the GLY to PG
20
21
22 conversion it was reported that with 80% aqueous glycerol, 200 °C and 200 psi of H2, a Raney
23
24 Ni catalyst gave a selectivity of 53% for PG at 50% conversion after 24 h. Under the same
25
26
conditions a Raney Cu catalyst yielded 69% selectivity to PG at 49% conversion. Some ACT
27
28
29 was also produced.33 Interestingly, no EG formation was reported in these articles even though
30
31 EG was the primary by-product observed in reactions run over supported Ni catalysts.15-17 An
32
33
34 early paper describing GLY conversion over Raney Cu at 240 °C and 750 psi of H2 reported
35
36 an 80% yield of PG and 20% EG at 85% conversion.43
37
38
39
40
Most of what has been published in the open literature described the use of batch reactions for
41
42 the GLY to PG reaction. However, continuous fixed bed processes are preferred over batch
43
44 procedures because the former provide a higher catalyst/feed ratio which offsets the low specific
45
46
47 activity of catalysts, such as those containing Cu, as well as having the ability to remove by-
48
49 products such as water and CO2 and, thus, minimizing any equilibrium constraints which may
50
51 have led to the low yields observed in most batch reactions. We report here an examination of
52
53
54 the effect which reaction variables have on the activity and selectivity of these Raney catalysts
55
56 in the continuous conversion of GLY to PG. A preliminary report on the use of Raney Cu for
57
58 6
59
60
1
2
3
the valorization of glycerol has been published.44 The data cited in this previous report have been
4
5
6 modified in the present description by the use of an improved analytical procedure involving five
7
8 point curves for the internal standard calibrations.
9
10
11
12 2. RESULTS AND DISCUSSION
13
14
15  Ni Catalysts. Relatively little has been published on the use of either Raney
2.1. Raney
16
17
18 Ni33,41,42 or Raney Cu33,43,44 for the conversion of GLY to PG. This seeming discrepancy
19
20 indicated that a re-investigation of the use of Raney Ni and Raney Cu catalysts in this
21
22
23
reaction was warranted but, given the reported relative inactivity of these catalysts in batch
24
25 processes33,41-43 as well as the advantages of fixed bed reactions cited above, the Raney Ni and
26
27
28
29 80
30
31 Conversion
32 PG Selectivity
33 EG Selectivity
EtOH Selectivity
34 60 MeOH Selectivity
35
36
Percent
37
38
40
39
40
41
42
43 20
44
45
46
47
48
49 180 185 190 195 200 205 210
50
Temperature, °C
51
52
53 Figure 1. Effect of reaction temperature on the percent conversion and product
54 composition for the GLY to PG reaction run over Raney Ni with an 80% aqueous
55 GLY solution flow rate of 0.075 cc/min and a H2 pressure of 300 psig and a flow
56
57
rate of 150 cc/min.
58 7
59
60
1
2
3
4
5
6
40
7
8
9
10
11 30
Percent
12
13 Conversion
PG Selectivity
14 EG Selectivity
15 20 EtOH Selectivity
16 MeOH Selectivity
17
18
19 10
20
21
22
23
24 200 300 400 500 600
25
26 Pressure, psi
27
28 Figure 2. Effect of H2 pressure on the percent conversion and product composition
29
30 for the GLY to PG reaction run over Raney Ni at 190 °C, with an 80% aqueous
31 GLY solution flow rate of 0.075 cc/min and a H2 flow rate of 150 cc/min.
32
33
34 Raney Cu promoted reactions were run in a continuous mode using a fixed bed reactor.
35
36
37
Research was initiated by an examination of the effect which the individual reaction parameters,
38
39
40 temperature, H2 pressure, GLY feed flow rate and H2 flow rate, had on the outcome of the
41
42 Raney Ni catalyzed reaction. Figure 1 and Figure 2 depict the effect of temperature and
43
44
45 hydrogen pressure, respectively, on the GLY conversion and reaction selectivity while the effect
46
47 of hydrogen and GLY feed flow rates are shown in Table 1. As Scheme 1 illustrates, the overall
48
49
reaction sequence for GLY conversion is complex with the potential of having several reactions
50
51
52 taking place simultaneously. In the equation at the top of Scheme 1, the dehydration of GLY
53
54 followed by hydrogenation of ACT, leads to the production of PG. The pathways at the bottom
55
56
57
58 8
59
60
1
2
3
of Scheme 1 are responsible for the degradation of GLY to EG, EtOH and MeOH. The totality
4
5
6 of these reactions is indicated by the percent conversion of GLY with the most significant effect
7
8 on this aspect of the reaction being brought about by changes in the reaction temperature since
9
10
11
this factor influences virtually all the reactions in Scheme 1. Increasing the hydrogen pressure,
12
13 however, had a more modest effect on the percent conversion since not all of the reactions used
14
15 hydrogen. A faster GLY flow rate can cause a decrease in conversion because the reactants have
16
17
18 insufficient time in contact with the catalyst to react completely.
19
20
21 The effect of these variables on the product composition, though, is more complex. A low GLY
22
23 feed flow rate resulted in the reaction mixture being in contact with the catalyst longer, thus,
24
25
26 giving more degradation products. Increasing the hydrogen flow rate also increased the extent of
27
28 degradation since there was more hydrogen available for hydrogenolysis. An increase in
29
30
31
temperature apparently favored the hydrogenation steps over the C-C hydrogenolysis while
32
33 increasing the hydrogen pressure appeared to have the opposite effect. The n-PrOH and ACT
34
35 selectivities were less than 1% and the 1,3PG selectivity was between 1% and 1.5% in all of
36
37
38
39
40 Table 1. Effect of H2 and GLY feed flow rates on percent conversion
41
and selectivity of PG, EG, EtOH and MeOH in a Raney Ni
42
43 catalyzed conversion of GLY.a
44 H2 Flow, 100 150 200 100 150 200
45 cc/min
46 GLY Feed 0.025 0.025 0.025 0.075 0.075 0.075
47 Flow, cc/min
48
49
50
% Conversion 77 75.4 73 46 39.2 38
51 % Selectivity
52 PG 17.0 17.8 20.7 18.8 20.8 22.7
53 EG 20.0 24 32.3 25.1 34.6 39.5
54 EtOH 14.9 13.5 11.1 7.1 7.0 7.7
55 MeOH 7.6 7.4 6.1 2.8 3.2 3.6
56 a
Temp. 190 °C, H2 pressure, 600 psi, 80% aqueous GLY.
57
58 9
59
60
1
2
3
these reactions.
4
5
6
7  Cu Catalysts. The large amount of EG, EtOH and MeOH formed in the Raney
2.2. Raney
8
9 Ni promoted reactions showed that it was not a viable catalyst for the selective conversion of
10
11
12 GLY to PG. It seemed, therefore, that since the use of other Raney type catalysts had not been
13
14 explored to any great extent primarily because of their commercial rarity and obscurity to date, a
15
16
17 non-Ni catalyst would be a prime candidate for further study especially when used in continuous
18
19 reactions. Given the interest in the use of Cu containing catalysts for the GLY to PG
20
21
conversion,19-38 it was decided to investigate the use of Raney Cu and modified Raney Cu
22
23
24 catalysts for this reaction.
25
26
27 The ranges of the reaction parameters investigated in the Raney Cu promoted reactions were
28
29
30 based on the data obtained with Raney Ni. The effect of temperature in the range of 190 °C to
31
32 205 °C on the conversion and product composition in the Raney Cu catalyzed reactions are
33
34
35 shown in Table 2. As expected, increasing the
36 Table 2. Temperature effect on Raney Cu
37 temperature increased the conversion and the catalyzed conversion of GLY to PG.a
38
39
40
yield of PG. The amount of EG observed Temp. % % PG % PG % EG
41 °C Conv.b Yield Sel.c Sel.c
42 was significantly less than that formed during
43 190 86.3 73.2 84.9 2.6
44 the Raney Ni catalyzed reactions. Table 3
45 200 99.9 80.5 80.6 2.3
46
47 lists the percent conversion and the PG and
48
205 100 81.4 81.4 1.7
49 EG selectivities observed from reactions run a
50 Reactions run at a H2 flow rate of 150 cc/min
51
with varying GLY feed and H2 flow rates as and pressure of 200 psi and 80% aqueous GLY
52 at a flow rate of 0.05 cc/min.
53
b
54 well as H2 pressures. In all reactions the Conversion
55 c
56
Selectivity
57
58 10
59
60
1
2
3
Table 3. Effect of GLY feed flow rate and H2 flow rate and pressure on the
4
5 Raney Cu catalyzed conversion of GLY to PG.a
6 GLY Feed H2 Flow H2
% % PG % EG
7 Flow Rate Rate Pressure
8 Conversion Selectivity Selectivity
cc/min cc/min psi
9
10
11 0.05 375 200 100 96.3 1.5
12
13 0.06 375 200 100 95.3 2.3
14
15 0.075 375 200 100 93.9 1.9
16
17 0.10 375 200 99.1 93.7 2.2
18
19 0.10 375 175 98.6 90.1 2.1
20
21
0.10 375 150 97.0 89.3 1.9
22
23
24 0.10 450 175 98.7 90.2 1.9
25
26 0.10 450 150 98.5 88.9 1.9
27
28 0.125 450 150 94.9 85.8 1.9
29
30 0.125 515 150 95.5 85.3 2.0
31
32 0.125 515 125 95.1 83.5 1.7
33
34 a
Reactions run using an 80% aqueous GLY at 205 °C.
35
36
37
38
39 amounts of MeOH, EtOH and ACT were less than 1% with no n-PrOH or1,3PG observed.
40
41 Increasing the GLY feed flow rate, with the concomitant decrease in residence time in the
42
43 catalyst bed, resulted in a decrease in PG selectivity regardless of the H2 flow rate. Increasing
44
45
46 the GLY feed flow rates above 0.1 cc/min decreased the percent conversion as well. As
47
48 anticipated from the reactions shown in Scheme 1, decreasing the hydrogen pressure resulted in
49
50 lower GLY conversions and PG formations. A higher H2 flow rate somewhat compensated for a
51
52
53 lower pressure. The data in Table 4 show that increasing the GLY concentration to 90% resulted
54
55 in a decrease in PG selectivity. Since virtually all of the previous research cited in the
56
57
58 11
59
60
1
2
3
Introduction used aqueous solutions of GLY it Table 4. Effect of glycerol concentration on
4
5 the Raney Cu catalyzed conversion of
6 would appear that the presence of some water GLY to PG.a
7
8 may be needed for this reaction. One possible GLY % % PG % EG
9
Conc.b Conv.c Sel.d Sel.d
10
11
scenario is that sufficient water in the GLY
12 80% 100 96.7 1.1
13 solution is needed to facilitate the passage of
14 90% 100 91.4 1.5
15 the substrate and hydrogen through the liquid- a
16 Reactions run at 205 °C with a H2 flow rate
17 of 375 cc/min and pressure of 200 psi and a
18 solid barrier at the catalyst surface. If this is
GLY feed flow rate of 0.05 cc/min.
19 b
20 the case then it would appear that the 90% Concentration.
c
21 Conversion.
22 GLY does not contain sufficient water for this d
Selectivity.
23
24
25 to happen.
26
27
28 Since the addition of Cr has been shown to increase the activity and selectivity of supported Cu
29
30
31 catalysts,25 some Cr modified Raney Cu catalysts were also used to promote the GLY to PG
32
33 reaction. They were Raney Cu containing 2% Cr and Raney Cu containing 0.75% Cr and
34
35
36 2.8% Ni. These catalysts along with a commercial CuCrO catalyst and the unmodified Raney
37
38 Cu were used in long term GLY to PG reactions with the results shown in Table 5. While all of
39
40
41
the reactions went to complete conversion, the use of the unmodified Raney Cu catalyst
42
43 resulted in the highest PG selectivity. The drop in PG selectivity using Raney Cu-Cr suggests
44
45
that the final hydrogenation step shown in Scheme 1 takes place primarily on the Cu surface.
46
47
48 With the Raney Cu-Cr catalyst, the Cu is probably partially covered by CrOx moieties. This
49
50 decrease in hydrogenation ability allows time for the competing degradation reactions to take
51
52
53 place. The Cu-CrO catalyst gave a differing product distribution, less PG and more EtOH. In
54
55
56
57
58 12
59
60
1
2
3
the first days of operation with this catalyst the product mixture was strongly colored suggesting
4
5
6 Table 5. Percent conversions and product compositions from GLY hydrogenolysis over copper
7 catalysts.a
8
9 % Selectivities
10
11
TOSb % n-
12 Catalyst PG EG MeOH EtOH ACT 1,3PG
13 (Days) Conv. PrOH
14
15 Raney 6.2 100 95.0 1.9 0.9 0 0.2 0.4 0
16 Cu
17 Raney 4.9 100 87.4 0.6 1.0 3.9 0 0.5 1.0
18
19
Cu-Cr
20 Raney 3.9 100 82.9 1.8 0.2 0 0.2 0.4 0
21 Cu-Cr-Ni
22 Cu-CrO 5.0 100 82.5 0.7 0.9 2.2 0 0.5 0.5
23
24 a
Reactions run at 205 °C with a H2 flow rate of 375 cc/min and pressure of 200 psi and an 80%
25
26
GLY feed flow rate of 0.05 cc/min.
b
27 Time on Stream.
28
29
30
31
32
either some leaching or the presence of colored unanalyzed oligomers. This later is more
33
34 consistent with the lower material balance observed with Cu-CrO (~92%) as opposed with
35
36 98.5% with Raney Cu. Since there was no interest in pursuing the use of Cu-CrO or the other
37
38
39 modified catalysts, this matter was not investigated further.
40
41
42 The unmodified Raney Cu was stable under these same reaction conditions in runs of up to
43
44
45 24 d with the conversion remaining at 100% and a stable PG selectivity at 95%. However, under
46
47 these reaction conditions the STY was only 0.17 grams of PG/cc of catalyst/h. Data in the
48
49 literature on the use of supported Cu catalysts show that higher temperatures could be used with
50
51
52 Cu catalysts but this was usually associated with higher H2 pressures because the increase in
53
54 temperature would decrease the H2 coverage of the catalyst.19-21,25 With this in mind, the effect
55
56
57
of increasing the reaction temperature was examined using a 50% aqueous GLY solution. The
58 13
59
60
1
2
3
data in Table 6 show that increasing the Table 6. Effect of higher temperatures on
4
5 the Raney Cu catalyzed conversion
6 temperature to 240 °C had no effect on the of GLY to PG.a
7 Temperature % % PG
8 GLY conversion. The PG selectivity was °C Conversion Selectivity
9
10
stable between 210 °C and 225 °C but 210 100 96.2
11
12
13 decreased at the higher temperature. It was 225 100 96.3
14
15 240 100 83.8
decided that further reactions would be run
16
a
17 Reactions run with a H2 flow rate of 300 cc/min
18 at 225 °C with the H2 pressure increased to
and pressure of 600 psi and a GLY feed flow rate
19
20 600 psi and a H2 flow of 300 cc/min. of 0.10 cc/min.
21
22
23 With these changes the effects of GLY feed concentration and flow rate would have to be
24
25
26 reexamined. In comparing the effect of different GLY feed concentrations, though, the flow rate
27
28 is not a useful parameter since the use of different concentrations of GLY feed at the same flow
29
30
31
rate would result in differing amounts of GLY feed passing through the catalyst over a given
32
33 time. The factor which should be used is the Weight Hourly Space Velocity (WHSV) which is
34
35 defined as the grams of substrate/gram of catalyst/h. The results from reactions run using 50%,
36
37
38 65% and 80% aqueous GLY feeds at different WHSV’s are shown in Figures 3 and 4. These
39
40 data indicate that there is a combination of factors involved in the GLY to PG reaction. Among
41
42 these is the effect which increasing the water concentration has on the competitive adsorption on
43
44
45 the catalyst surface between water and GLY. Another factor is the changing of the contact time
46
47 (flow rate or WHSV) of the reactants with the catalyst and the effect it has on the kinetics of the
48
49
reaction. A third factor is the effect which increasing the water concentration had on the
50
51
52 reactions shown in Scheme 1. There are two separate aspects of this reaction which are
53
54 summarized in Figures 3 and 4, the percent conversion of GLY and the selectivity in the
55
56
57
formation of PG and EG from the GLY.
58 14
59
60
1
2
3
4
5
6
7
8
100
9
10
Percent Conversion
11
12
13
95
14
15
16
17
50% Glycerol
18 65% Glycerol
90
19 80% Glycerol
20
21
22
23 85
24
25 0.2 0.3 0.4 0.5 0.6 0.7
26
27 WHSV
28
29
Figure 3. WHSV effect on the percent conversion observed using different aqueous GLY
30
31 concentrations for the GLY to PG reaction run over Raney Cu at 225 °C
32 with a H2 pressure of 600 psig and a flow rate of 300 cc/min.
33
34
35 Increasing the flow rate will decrease the contact time of the water and GLY on the catalyst
36
37
38
surface. This fact, combined with the competitive adsorption of water on the Cu surface, would
39
40 indicate that at a particular flow rate GLY adsorption could be diminished to the extent that the
41
42 conversion of GLY decreased. As shown in Figure 3, this point was reached with the 50%
43
44
45 aqueous GLY at a flow rate of about 0.15 cc/min (WHSV = 0.35) while with the 65% GLY it
46
47 was at a flow rate of about 0.19 cc/min (WHSV = 0.57) and with 80% GLY about 0.15 cc/min
48
49 (WHSV = 0.55). This was not an indication of catalyst deactivation since when the flow rate
50
51
52 was decreased the conversion returned to 100%. This type of behavior was observed at other
53
54 times when changes in the GLY feed flow rate was being examined. When a decrease in GLY
55
56
57
58 15
59
60
1
2
3
conversion was noted with an increase in the GLY feed flow rate the conversion could be
4
5
6 recovered by returning to the original flow rate.
7
8
9
10 100
11
12
13
14
Percent Selectivity
15 PG
16 50% Glycerol
17
65% Glycerol
18
19 80 80% Glycerol
20
21 8
22 EG
23 6
24
25 4
26
27
2
28
29
30
31 0.2 0.3 0.4 0.5 0.6 0.7
32 WHSV
33
34
35 Figure 4. WHSV effect on the PG and EG selectivities observed using different
36 aqueous GLY concentrations for the GLY to PG reaction run over Raney Cu at
37 225°C with a H2 pressure of 600 psig and a flow rate of 300 cc/min.
38
39
40 The product compositions shown in Figure 4, though, appear to depend on the effect which water
41
42
43 had on the different reaction pathways depicted in Scheme 1. It is possible that an increase in the
44
45 water content of the feed could inhibit the initial dehydration of GLY to form ACT and this
46
47
48
would decrease the formation of PG while shifting the catalytic activity to those paths which
49
50 lead to the formation of EG, pathways in which water does not play a role. As the water content
51
52 of the feed decreased, the amount of PG increased, though slightly, while the EG selectivity
53
54
55 decreased significantly.
56
57
58 16
59
60
1
2
3
Table 7. Product composition from 24 d (580 h) continuous GLY hydrogenolysis
4
5 catalyzed by Raney Cu.a
6 % Selectivities
7
8 % Conv. PG EG MeOH EtOH n-PrOH ACT
9
10 100 94.6 2.5 0.4 0.7 0.6 1.0
11
12 a
Reaction run at 225 °C with a H2 flow rate of 300 cc/min and pressure of 600 psi and
13
14 an 80% GLY flow rate of 0.12 cc/min.
15
16
17
18
19 In Table 7 are shown the average product compositions for a reaction run continuously for 24 d
20
21
22 using 80% aqueous GLY over a Raney Cu catalyst at 225 °C, a H2 flow rate of 300 cc/min, a
23
24 H2 pressure of 600 psi and a GLY feed flow rate of 0.12 cc/min (WHSV = 0.440). Product
25
26
compositions varied by less than 1% throughout the reaction. The STY, under these conditions,
27
28
29 was 0.49 g PG/cc catalyst/h corresponding to 3.98 kg of PG being produced over the course of
30
31 this reaction. The catalyst used for the reaction was a sample obtained from a commercial scale
32
33
34 production of Raney Cu.
35
36
37 3. CONCLUSION
38
39
40
41
A commercially prepared Raney Cu catalyst has been shown to be an efficient and selective
42
43 catalyst for the conversion of glycerol to propylene glycol under moderate conditions; 80%
44
45 aqueous glycerol, 225 °C, H2 pressure of 600 psig in a liquid phase fixed bed reactor. The
46
47
48 unmodified Raney Cu catalyst was shown to be superior to Raney Νi, Cu-CrO and Raney
49
50 Cu catalysts containing Cr and Ni modifiers.
51
52
53
54 4. EXPERIMENTAL SECTION
55
56
57
58 17
59
60
1
2
3
4
4.1. Materials. The Raney catalysts used in this study were made using proprietary
5
6 techniques involving the partial leaching of Al from 8 to 12 mesh sized particles of Ni-Al, Cu-
7
8 Al, Cu-Al-Cr and Cu-Al-Ni-Cr alloys using dilute NaOH solutions. The copper chromite (Cu-
9
10
11 CrO) catalyst was a 1/8”extrudate obtained from BASF, Elyria, OH. The sample used was
12
13 obtained by first crushing the extrudates and sieving the resulting material to 10-20 mesh before
14
15
packing the catalyst in the reactor. The Cu-CrO catalyst was reduced, in situ, in a 20 mL/min
16
17
18 flow of hydrogen at 205 °C for 2 h, after raising the temperature at 20°/h from ambient to 205
19
20 °C. The Raney catalysts did not require any in situ activation.
21
22
23
24 4.2. Description of the fixed bed reactor system. The continuous hydrogenation of glycerin
25
26 over particulate Raney catalysts was carried out in the fixed bed reactor depicted in Figure 5. The
27
28
gas control section, A, consisted of a channel for low pressure inert gas, a low pressure hydrogen
29
30
31 line, a by-pass channel for pressurizing the system and a high pressure hydrogen delivery
32
33 channel. All four channels were fitted with Brooks mass flow controllers (6) and high pressure
34
35
36
two-way (7) and three-way (8) solenoid valves (Parker Instruments). All controllers and the
37
38 solenoid valves were interfaced to an in-house developed data acquisition and control system.
39
40
41 Section B was the GLY feed unit. The GLY solution, in a glass reservoir (9), was kept under 30
42
43
44 psi nitrogen pressure and placed on a balance for continuous monitoring and recording of the
45
46 amount of feed added over time. The solution was pumped into the reactor using a high precision
47
48 Shimadzu LC-10AT pump (10) while the liquid flow rate was also monitored by a Brooks liquid
49
50
51 mass flow meter (11).
52
53
54 Section C was a 0.5 inch id tube reactor fitted with three band-heaters and quarter inch inlet and
55
56
57
outlet ports. The reactor was partitioned into three zones with the temperatures in each zone
58 18
59
60
1
2
3
controlled by two Omega temperature controllers; one for the internal reactor temperature and
4
5
6 the second for controlling the temperature of the band heater. This combination of two
7
8
A PC 6-1 B
9
10 Aqueous Glycerol
11 HP LP
12 7-1 PC LP
Argon 9
13 6-2
14
15 LP
16 PC PC LP
17 Pump 10
18 HP PC
19 7-2
20 PC 6-3
8-1 MFM 11
21 Hydrogen
22 8-2
23 PC HP Inlet 8-3
24 Pressure
25 PC
26 PC
27
28 PT PC
29 13-1
30 TC-1
31
C
32
33 TC-3 TC-4
34 Outlet 12
35 PC PC
Pressure PC
36
37
Gaseous product PT
38 D
39 BPR Condenser 13-2 12
40 TC-5
Sample
41 Collector
42
15 14 TC-2 PC
43 16
44 Liquid
45 Product
46
47
48 Figure 5. Fixed bed reactor
49
50 controllers connected in series enabled maintaining the temperature of each zone within ±0.2 °C.
51
52
53 The top, pre-heating, zone was fitted with 2 µm stainless steel filters, 12, and used for pre-
54
55 heating and mixing the aqueous glycerin and hydrogen. The second zone was the “hot catalyst”
56
57
58 19
59
60
1
2
3
temperature zone with a thermocouple inserted 0.5 inch into the catalyst bed. The third zone was
4
5
6 the catalyst bed temperature zone with a second thermocouple placed 0.5 inch into the lower
7
8 section of the catalyst bed. The reactor inlet and outlet pressure were monitored by Omega
9
10
11
pressure transducers (13).
12
13
14 The fourth section,D, consisted of a condenser (14), a domed back pressure regulator (15) and a
15
16 gas–liquid separator (16) operated at a temperature of 6 oC. The excess hydrogen flow was
17
18
19 directed to a trap kept in dry ice.
20
21
22 All two and three way valves, the mass flow controllers and meters, the reactor temperatures, the
23
24
inlet and outlet pressures and the balance reading were constantly under computer control and
25
26
27 monitoring. Special attention was paid to the safety of the flow system so when any one of the
28
29 process parameters (temperatures in the reactor, pressure, gas and liquid flows) varied by more
30
31
32
than 10% from the pre-set values, the system was programmed to enter an “abort regime” which
33
34 terminated the power to the heaters and the pump and vented the reactor to a by-pass line.
35
36
37 4.3 Hydrogenolysis of GLY. Three filtering elements, 12, were inserted over the bottom
38
39
40 thermocouple to serve as a catalyst support (Figure 5). The catalyst, 14 cc (16.5 g), was
41
42 suspended in water and carefully transferred into the reactor under a constant flow of water for
43
44 uniform packing and protection from direct contact with air. A GC vibrator was used to avoid
45
46
47 gaps between the particles. The reactor was sealed on the top with three additional filtering
48
49 elements, 12, to contain the catalyst bed. The reactor was attached to the feed and the exit lines.
50
51
52
53
The GLY solutions were prepared using de-ionized water at pH 7. At each set of reaction
54
55 conditions the catalyst performance was stabilized by running the reaction for 6-7 h after which
56
57
58 20
59
60
1
2
3
time 3 collections were made; 2 day-time collections for an average TOS of 8 h each and one
4
5
6 overnight collection with an approximate 15 h TOS. All reactions were run in a down flow
7
8 mode with both the aqueous GLY and the hydrogen introduced at the top of the reactor. The
9
10
11
product flow exited the reactor, passed through a condenser, 14, which was kept at 25 °C, and
12
13 then to the BPR, 15, which lowered the pressure to ambient. The exit flow was switched to a
14
15 waste flask during the stabilization steps or to a collection flask during the reaction steps and
16
17
18 analyzed offline. The gas phase was passed through a liquid trap held at -36 °C to condense low
19
20 boiling products which may have been in the hydrogen flow.
21
22
23 4.4 GC analysis was done using an HP 5890 with FI detector and an RTX1701 capillary
24
25
26 column (60 m, 0,53 mm ID and film thickness of 1 µm). 1,4-Butanediol was used as the internal
27
28 standard with the response factor (RF) calibrations for the reaction products obtained using 5
29
30
31 point curves.
32
33
34 FORMULAS
35
36 % Conversion of GLY = [mmol converted GLY] * 100 / [mmol GLY]o
37
38
39
% Selectivity to PRODUCT = [mmol PRODUCT formed] * 100 / [mmol converted GLY]
40
41 Space-Time-Yield (STY) = g PRODUCT/cc catalyst/h
42
43 Weight Hourly Space Velocity (WHSV) = g GLY/g catalyst/h
44
45 Liquid Hourly Space Velocity (LHSV) = cc GLY/cc catalyst/h
46
47
48
49
50 REFERENCES
51
52
53 (1) Stelmachowski, M. Ecological Chemistry and Engineering 2011, 18, 9-30.
54
55
56
57
58 21
59
60
1
2
3
(2) Katryniok, B.; Kimura, H.; Skrzynska, E.; Girardon, J.S.; Fongarland, P.; Capron, M.;
4
5
6 Ducoulombier, R.; Mimura, N.; Paul, S.; Dumelgnil, F. Green Chem. 2011, 13, 1960-1979.
7
8
9 (3) Suppes, G.J. The Biodiesel Handbook, Second Edition, Knothe, G., Krahl, J. and Van
10
11
12 Gerpen, J. eds. AOCS Press, Urbana, IL, 2010, Chapter 11.
13
14
15 (4) Nakagawa, Y.; Tomishige, K. Catalysis Science & Technology 2011, 1, 179-190.
16
17
18 (5) Zhou, C.-H.; Beltramini, J.N.; Fan, Y.-X.; Lu, G.Q. Chemical Society Reviews 2008, 37,
19
20
21 527-549.
22
23
24 (6) Martin, A.; Armbruster, U; Atia, H. Eur. J. Lipid Sci. Technol. 2012, 114, 10-23.
25
26
27
28
(7) Stelmachowski, M. Advances in Chemistry Research 2011, 9, 59-92.
29
30
31 (8) Ochoa-Comez, J.R.; Gomez-Jimenez-Aberastury, O.; Ramirez-Lopez, C.; Belsue, M. Org.
32
33 Process Res. Dev. 2012, 16, 389-399.
34
35
36
37 (9) Brandner, A.; Lehnert, K.;Bienholz, A.; Lucas, M.; Claus, P. Top. Catal. 2009, 52, 278-
38
39 287.
40
41
42
(10) Kusunoki, Y.; Miyazawa, T.; Kunimori, K.; Tomishige, K. Catalysis Communications
43
44
45 2005, 6, 645-649.
46
47
48 (11) Miyazawa, T.; Kusunoki, Y.; Kunimori, K.; Tomishige, K. J. Catal. 2006, 240, 213-221.
49
50
51
52 (12) Miyazawa, T.; Koso, S.; Kunimori, K.; Tomishige, K. Appl. Catal. A: General, 2007, 329,
53
54 30-35.
55
56
57
58 22
59
60
1
2
3
(13) Maris, E.P.; Ketchie, W.C.; Murayama, M.; Davis, R.J. J. Catal. 2007, 251, 281-294.
4
5
6
7 (14) Li, B.; Wang, J.; Yuan, Y.; Ariga, H.; Takakusagi, S.; Asakura, K. ACS Catal. 2011, 1,
8
9 1521-1528.
10
11
12
13 (15) Yu, W.; Xu, J.; Ma, H.; Chen, C.; Zhao, J.; Miao, H.; Song, Q. Catalysis Communications
14
15 2010, 11, 493-497.
16
17
18 (16) Yu, W.; Zhao, J.; Ma, H.; Miao, H.; Song, Q.; Xu, J. Appl. Catal. (A): General 2010, 383,
19
20
21 73-78.
22
23
24 (17) Zhao, J.; Yu, W.; Chen, C.; Miao, H.; Ma, H.; Xu, J. Catal. Lett. 2010, 134, 184-189.
25
26
27
28
(18) Guo, X.; Li, Y.; Shi, R.; Liu, Q.; Zhan, E.; Shen, W. Appl. Catal. (A): General 2009, 371,
29
30 108-113.
31
32
33 (19) Tuck, M.W.M.; Tilley, S.N. US Patent 7,355,083, Apr. 8, 2008, to Davy Process
34
35
36 Technology Limited, London.
37
38
39 (20) Casale, B.; Gomez, A.M. US Patent 5,214,219 May 25, 1993, to Novamont S.p.A., Milan.
40
41
42
(21) Wang, S.; Liu, H. Catalysis Letters 2007, 117, 62-67.
43
44
45
46 (22) Bienholz, A.; Schwab, F.; Claus, P. Green Chem. 2010, 12, 290-295.
47
48
49 (23) Franke, O.; Stankowiak, A. US Patent 7,812,200 Oct. 12, 2010 to Clariant Finance (BVI)
50
51
52 Limited, Tortola.
53
54
55
56
57
58 23
59
60
1
2
3
(24) Kuono, H.; Ozawa, S.; Naritoshi, Y. US Patent 8,053,608, Nov. 8, 2011 to Mitsui
4
5
6 Chemicals Inc., Tokyo.
7
8
9 (25) Kim. N.D.; Oh, S.; Joo, J.B.; Jung, K.S.; Yi, J. Top. Catal. 2010, 53, 517-522.
10
11
12
13 (26) Liang, C.; Ma, Z.; Ding, L.; Qiu, J. Catal. Lett. 2009, 130, 169-176.
14
15
16 (27) Xiao, Z.; Ma, Z.; Wang, X.; Williams, C.T.; Liang, C. Ind. Eng. Chem. Res. 2011, 50,
17
18 2031-2039.
19
20
21
22 (28) Ma, Z.; Xiao, Z.; van Bokhoven, J.A.; Liang, C. J. Mater. Chem. 2010, 20, 755-760.
23
24
25 (29) Xiao, Z.; Li, C.; Ziu, J.; Wang, X.; Williams, C.T.; Liang, C. J. Mol. Catal. A: Chemical
26
27
28
2012, 365, 24-31.
29
30
31 (30) Meher, L.C.; Gopinath, R.; Naik, S.N.; Dalai, A.K. Ind. Eng. Chem. Res. 2009, 48, 1840-
32
33 1846.
34
35
36
37 (31) Schuster, L.; Eggersdorfer, M. US Patent 5,616,817, Apr. 1, 1997 to BASF
38
39 Aktiengesellschaft, Ludwigshafen, Germany.
40
41
42
(32) Stankowiak, A.; Franke, O. US Patent 7,868,212,B2, Jan 11, 2011 to Clariant Finance
43
44
45 (BVI) Limited, Tortola.
46
47
48 (33) Desari, M.A.; Kiatsimkul, P.-P.; Sutterlin, W.R.; Suppes, G.J. Appl. Catal. A: General
49
50
51
2005, 281, 225-231.
52
53
54 (34) Chiu, C.-W.; Tekeei, A.; Ronco, J.M.; Banks, M.-L.; Suppes, G.L. Ind. Eng. Chem. Res.
55
56 2008, 47, 6878-6884.
57
58 24
59
60
1
2
3
(35) Suppes, G.L.; Sutterlin, W.R.; Dasari, M. US Patent 7,816,567, Oct. 19, 2010 to The
4
5
6 Curators of the University of Missouri, Columbia, MO and Renewable Alternatives LLC,
7
8 Columbia, MO.
9
10
11
12 (36) Suppes, G.L.; Sutterlin, W.R.; Dasari, M. US Patent 7,943,805, May 17, 2011 to The
13
14 Curators of the University of Missouri, Columbia, MO.
15
16
17 (37) Suppes, G.L.; Sutterlin, W.R.; US Patent 8,017,816, Sept. 13, 2011 to The Curators of the
18
19
20 University of Missouri, Columbia, MO and Renewable Alternatives LLC, Tuscaloosa, AL.
21
22
23 (38) Suppes, G.L. US Patent 8,252,961, Aug. 28, 2012 to The Curators of the University of
24
25
Missouri, Columbia, MO.
26
27
28
29 (39) Chiu, C.-W.; Dasari, M.A.; Suppes, G.L.; Sutterlin, W.R. AIChE Journal 2006, 52, 3543-
30
31 3548.
32
33
34
35 (40) Chiu, C.-W.; Tekeei, A.; Sutterlin, W.R.; Ronco, J.M.; Suppes, G.L. AIChE Journal 2008,
36
37 54, 2456-2463.
38
39
40 (41) Hosgun, H.L.; Yildiz, M.; Gercel, H.F. I&EC Research 2012, 51, 3863-3869.
41
42
43
44 (42) Perosa, A.; Tundo, P. Ind. Eng. Chem. Res. 2005, 44, 8535-8537.
45
46
47 (43) Montassier, C.; Giraud, D.; Barbier, J. Stud. Surf. Sci. Catal. 1988, 41 (Heterogeneous
48
49
50
Catalysis and Fine Chemicals), 165-170.
51
52
53 (44) Schmidt, S.R.; Tanielyan, S.K.; Marin, M.; Alvez, G.; Augustine, R.L. Top. Catal. 2010,
54
55 53, 1214-1216.
56
57
58 25
59
60

An Efficient, Selective Process For The Conversion of Glycerol To Propylene Glycol Using Fixed Bed Raney Copper Catalysts

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

An Efficient, Selective Process For The Conversion of Glycerol To Propylene Glycol Using Fixed Bed Raney Copper Catalysts

Uploaded by

Copyright:

Available Formats

Subscriber access provided by Lulea University of Technology

Organic Process Research & Development is published by the American Chemical

You might also like