Integration of phosphine ligands and ionic liquids both in structure and properties-a new strategy for separation, recovery, and recycling of homogeneous catalyst

View Article Online

View Journal

Green

Chemistry

Accepted Manuscript

This article can be cited before page numbers have been issued, to do this please use: X. Jin, J. Feng, H.

Song, J. Yao, Q. Ma, M. Zhang, C. Yu, S. Li and S. Yu, Green Chem., 2019, DOI: 10.1039/C9GC00679F.

This is an Accepted Manuscript, which has been through the

Volume 18 Number

7 7 April 2016 Pages 1821–2242

Royal Society of Chemistry peer review process and has been

accepted for publication.

Green

Chemistry

Accepted Manuscripts are published online shortly after

Cutting-edge research for a greener sustainable future

www.rsc.org/greenchem

acceptance, before technical editing, formatting and proof reading.

Using this free service, authors can make their results available

to the community, in citable form, before we publish the edited

article. We will replace this Accepted Manuscript with the edited

and formatted Advance Article as soon as it is available.

You can find more information about Accepted Manuscripts in the

author guidelines .

Please note that technical editing may introduce minor changes

to the text and/or graphics, which may alter content. The journal’s

standard T erms & Conditions and the ethical guidelines, outlined

in our author and reviewer resource centre, still apply. In no

event shall the Royal Society of Chemistry be held responsible

for any errors or omissions in this Accepted Manuscript or any

consequences arising from the use of any information it contains.

ISSN 1463-9262

CRITICAL REVIEW

G. Chatel et al.

Heterogeneous catalytic oxidation for lignin valorization into valuable

chemicals: what results? What limitations? What trends?

rsc.li/green-chem

Page 1 of 14

Green Chemistry

View Article Online

DOI: 10.1039/C9GC00679F

RSCPublishing

Green Chemistry

ARTICLE

Integration of Phosphine Ligand and Ionic Liquid Both in Structure

and Property: An Efficient and Economical Catalytic System for

Homogeneous-Catalyst Recycling^†

Xin Jin,*^aJianying Feng,^aHongbing Song,^aJiajun Yao,^aQingqing Ma,^aMei Zhang,^aCong Yu,^aShumei

Li,^band Shitao Yu^a

The major limitation of classic biphasic ionic liquid (IL) catalysis is the heavy use of solvent IL, which not only violates green

chemistry principles but also even worsens catalytic efficiency. So it has always been a challenge how to use IL more

efficiently, economically, and greenly to construct highly effective and long term stable IL catalytic system. In this work, we

synthesized a class of room temperature phosphine-functionalized polyether guanidinium ionic liquids (RTP-PolyGILs) by a

convenient ion exchange reaction of polyether guanidinium ionic liquids (PolyGILs) with phosphine-sulfonate ligands based

on the concept of the integration of both phosphine ligand and IL. The resulting RTP-PolyGILs existed as liquids at room

temperature and possessed dual functions of both phosphine ligand and solvent IL, so they could both form catalysts by

complexing with transition metals and act as catalyst carriers, thus achieving the integration of phosphine ligands with ILs

both in structure and properties. Based on the unique properties of these multi-functional integrated RTP-PolyGILs, we

constructed a highly effective homogeneous catalysis-biphasic separation (HCBS) system for Rh-catalyzed hydroformylation

of higher olefins using only a catalytic amount of RTP-PolyGILs (equivalent to 0.025–0.4 mol% of 1-alkenes). Our HCBS system

could be flexibly regulated with regard to catalytic performance (activity and linear selectivity) by changing the structure or

type of the sulfonated ligand anion on RTP-PolyGILs. Specifically, it presented the TOF value of 3000–26000 h⁻¹and linear

selectivity of 68%–98% (corresponding to the l/b ratio of 2.2–37.5) with a total turnover number (TTON) of 11000–45000

and an extremely low Rh leaching in only 0.02–0.4 ppm. Therefore, the HCBS system can effectively combine the advantages

of both homogeneous (high activity, and good selectivity) and biphasic catalysis (easy catalyst separation). We additionally

extended the application of the HCBS system to the hydrogenation of olefins to demonstrate the universality of the RTP-

PolyGILs in catalytic reactions.

Received 00th February 2019,

Accepted 00th February 2019

DOI: 10.1039/x0xx00000x

www.rsc.org/greenchem

chemistry principles²and is economically burdensome, and may

even worsen certain problems in catalytic reactions. For instance, the

15extensive use of solvent IL may impede the diffusion of substrates

Introduction

The use of ionic liquid (IL) has been considered one of the most

competitive means for the practical separation, recovery, and

recycling of expensive transition metal materials in homogeneous

catalysis¹. Compared with traditional organic solvents, the attractive

properties of IL include negligible vapor pressure, excellent thermal

and chemical stability, high conductivity, and good solubility for

transition metal catalysts. As a result, IL has been successfully used

as an alternative solvent in liquid/liquid biphasic catalysis.

due to high viscosity of IL, and the reaction tends to occur at the

IL/substrate interface, which can reduce catalytic efficiency since

most of the active catalysts in bulk phase of IL cannot participate in

the reaction³. Besides, the impurities in the solvent IL can be of an

20excessive amount that leads to the oxidation and decomposition of

the catalysts and decreases their stability⁴. So far, it remains

challenging to use the IL more efficiently, greenly and economically,

and it remains a matter of utmost urgency to prepare long term

stable IL catalytic system that is highly effective (high activity, good

25selectivity, and easy catalyst separation).

5

10 The major limitation of biphasic IL catalysis involves the use of

large amounts of solvent IL that is often needed to support and

dissolve the catalyst. The heavy use of solvent IL violates green

The use of supported ionic liquid catalysis provides one means to

solve the above problem by joining the solvent IL (providing excellent

solubility) and the support material (rendering high specific surface

area) together to significantly reduce the use of solvent IL and

30improve the mass transfer⁵. However, we approach the problem

alternatively by simply functionalizing the IL itself, since the

functionality of the IL can be tailored to address particular needs⁶. In

^a.State Key Laboratory Base for Eco-Chemical Engineering in College of Chemical

Engineering, Qingdao University of Science and Technology, 53 Zhengzhou Street,

Qingdao 266042, China. E-mail: jinx1971@163.com (X. Jin).

^b.College of Physical Education, Qingdao University of Science and Technology, 53

Zhengzhou Street, Qingdao 266042, China.

† Electronic Supplementary Information (ESI) available: See

DOI: 10.1039/x0xx00000x

Green Chemistry

Page 2 of 14

ARTICLE

Green Chemistry

our opinion, the key to preparing highly effective, long-term stable IL

catalytic systems is to create functionalized ionic liquid (FIL) capable

of effectively integrating the ligand (or catalyst) and the solvent IL by

making use of the intrinsic designer characteristics of IL, such that

the FIL can possess not only the capacity of the ligand to complex

with the metal but also the features of the solvent IL to serve as the

solvent and carrier of the catalyst. In this way, the catalytic system

will only require a catalytic amount of the FIL that plays the dual role

of the ligand and the solvent IL simultaneously, and eliminates the

View Article Online

DOI: 10.1039/C9GC00679F

5

10need for excessive solvent IL. The efficient, economical, and

environmentally-friendly use of IL can be thus achieved.

Although some previous studies tried to attach phosphine ligands

to cations⁷(imidazolium, pyridinium, phosphonium, guanidinium,

and other onium ions) by covalent⁸or electrovalent⁹bonding and

15indeed synthesized some phosphine-functionalized ionic liquids (P-

FILs), the primary purpose of these studies was not to reduce the use

of solvent IL but only to reduce the loss of metal catalysts by

improving the affinity of phosphine species to and increasing their

solubility in the solvent ILs. So despite reporting some successes in

20reducing the metal leaching, it was still inevitable in the catalytic

reactions to add large amounts of regular solvent IL, up to 30–6500

times (mol/mol) the amount of the metal catalyst (equivalent to 30–

350 mol % substrate)^8,9, to fully dissolve and support these P-FILs

because they generally had relatively high melting point, resembled

25ionic ligands, and did not demonstrate the solvent properties of

solvent IL at room temperature. As a result, the troubles mentioned

above have still remained in these catalytic systems using high

melting point P-FILs. The only attempt to eliminate solvent IL and use

just a catalytic amount of P-FIL was reported in one case on the

30continuous flow hydroformylation in supercritical fluid¹⁰, in which

the P-FIL and the Rh precursor were dissolved in the reaction

medium that was a stable state mixture of olefin and aldehyde, and

the flow rate of the olefin had to be strictly controlled to maintain a

steady state to prevent the catalyst from precipitating. That is to say,

35these so-called P-FILs only achieved the structural integration of

phosphine ligands with ILs, rather than the properties integration.

Therefore, the integration of phosphine ligands into the solvent IL

needs to be further studied to reduce the melting point of P-FIL and

thus make it highly practically “functional”, i.e., to serve both as the

40solvent and as the catalyst.

In recent years, polyether ionic liquids have received increasing

attention, because they combine the advantages of both polyether

and IL¹¹. The incorporation of polyether groups can significantly

reduce the melting point and viscosity of classic ILs. Many polyether

45ILs have been applied as solvents in organic synthesis and catalysis.

Recently, we also reported a polyether guanidinium ionic liquid

(PolyGIL, 1b in Scheme 1)¹², whose melting point was at room

temperature, and its application in the biphasic hydroformylation of

higher olefins. However, this system only gave TOF values up to <200

50h⁻¹, presumably because the large use of PolyGIL (equivalent to 28

mol% of the 1-alkene substrate) as the solvent IL impeded the mass

transport (Scheme 1).

Scheme 1 Integration of both phosphine ligand and IL, our previous and current

60strategies in recycling of homogeneous catalysts.

guanidinium cations into the anion of sulfonated phosphine ligands

would endow the resulting new sulfonated phosphine ligand with

the features of PolyGIL, such as lower melting point, liquefaction at

65room temperature, and good tunable solubility, thus merging the

phosphine ligand with IL. By exploiting the unique properties of this

multi-functional integrated IL, it would be possible to develop new

catalytic systems that possess high efficiency and good long term

stability.

70 In this article, we attempted to design and synthesize a series of

room temperature phosphine-functionalized polyether guanidinium

ionic liquids (RTP-PolyGILs, 2 and 3 in Scheme 1) by the ion exchange

of PolyGILs (Scheme 1, 1) with phosphine-sulfonate ligands.

Afterwards, we demonstrated the Rh-catalyzed hydroformylation of

75long chain alkenes, which was used as a model reaction, in a

homogeneous catalysis-biphasic separation (HCBS) system using

RTP-PolyGILs (Scheme 1). In the HCBS system only a catalytic amount

of RTP-PolyGILs (equivalent to 0.025–0.4 mol% of 1-alkenes) was

used and there was no additional solvent ILs. We investigated how

80the anions and cations of the RTP-PolyGILs affected the catalytic

activity, selectivity, and stability of the Rh catalysts. We then

extended the application of our system to the hydrogenation of

olefins. With the developed RTP-PolyGILs, we aim to devise greener,

With the unique properties of polyether ILs in mind and based on

our previous work, we conjectured that should phosphine-sulfonate

55ligand like TPPTS be joined with PolyGIL via ion exchange, an

innovative room temperature phosphine-functionalized polyether

guanidinium IL could be obtained. Introduction of the polyether

2 | Green Chemistry, 2019, 00, 1-3

This journal is © The Royal Society of Chemistry 2019

Page 3 of 14

Green Chemistry

ARTICLE

more economic, and more efficient catalytic systems that possess

high activity, good selectivity, as well as facile separation, recovery, polyether chains can reduce the melting point of resulting sulfonated

and recycling of catalyst, all at the same time maintaining excellent phosphine ligands, we first examined their melting point with

Since we were very concerned about whether the in_Vc_io_ewrp_Ao_rtr_ica_leti_Oon_nlio_nef

DOI: 10.1039/C9GC00679F

long term catalyst stability.

differential scanning calorimetry (DSC) after preparing the RTP-

60PolyGILs. As expected, using polyether guanidinium cations to

replace Na⁺successfully turned dystectic Na-typed sulfonated

phosphine ligands (T_m>300 °C)¹⁵into ionic liquids melting at room

5

Results and discussion

temperature (T_{m 2b}= 19.3°C; T_{m 2d}= 10.6°C; T_{m 3b}= 13.8°C; T_{m 3d}

=

Structural Design, Synthesis, Melting point, and Solubility of RTP-

PolyGILs. The structural design followed the commonly

acknowledged rule for FILs (i.e., “build for purpose”). Hence, RTP-

PolyGILs were conveniently prepared without intricate synthetic

10steps using both PolyGILs and available sulfonated phosphine ligands

via ion exchange as a modular assembly.

The integration of multiple functions was realized by combining

both the phosphine ligand (for its ability to bind with metals) and the

solvent IL (for its excellent dissolving capacity), such that a catalytic

15amount of RTP-PolyGIL should suffice to allow the catalytic reaction

to proceed and to separate, recover, and recycle the catalyst. Many

negative impacts from the excessive use of solvent IL in conventional

biphasic IL catalysis can thus be minimized, such as impeded mass

transport and catalyst degradation.

20 In addition, this modular assembly may allow universality in the

catalyst for the following reasons. Firstly, the sulfonated ligands,

introduced as the anions of the RTP-PolyGILs, are either

commercially available or can be readily synthesized and selected

carefully according to the type of the catalytic reaction to be

25performed. Secondly, the catalytic performance of the transition

metal catalysts (e.g., activity, chemoselectivity, regioselectivity, and

stereoselectivity) can be flexibly regulated by changing the type and

structure of the sulfonated ligands. Lastly, when requirements from

the application vary, the physicochemical properties of the RTP-

30PolyGILs (melting point, solubility, viscosity, thermal stability, etc.)

can also be easily adjusted by modifying substituents and/or

changing the number of ethylene oxide units (EO number) on the

onium cation. For example, it is well known that the solvent effect is

an important influencing factor in homogeneous catalysis, however,

35the sulfonated ligands are mainly used in the aqueous-organic

biphasic catalysis because they are soluble only in water, but

insoluble in most weak polar solvents (such as benzene, toluene,

halohydrocarbon, etc.) and some polar solvents (such as alcohols,

acetonitrile, THF, etc.), which has limited the scope of their

40application. By adjusting the balance between hydrophiles and

lipophiles in the polyether guanidinium cation, the RTP-PolyGILs can

become soluble in the abovementioned solvents to extend the scope

of using sulfonated ligands from the commonly seen aqueous-

organic biphasic catalysis to even homogeneous catalysis, thanks to

45the additional solubility introduced from the RTP-PolyGILs.

35.1°C) or unobservable melting temperature (2a, 2c, 3a, and 3c),

65which means that both phosphine ligands and solvent ILs had been

successfully unified as a whole not only in structure but also in

properties, so that RTP-PolyGILs alone can be used to support

catalysts without the need for solvent IL. This could not be achieved

by attaching the ordinary ionic groups to neutral ligands^8,9. The

70reduction in T_mwas attributed to the higher rotational freedom and

the greater flexibility of the ether chain, which decreased the lattice

energy¹⁶.

We next turned our attention to the solubility of the RTP-PolyGILs,

because their miscibility with reaction substrates or products and

75their solubility in various solvents are vital properties in catalytic

reactions, and will determine not only catalytic efficiency but also

under what system does the reaction proceed. Homogeneous or

biphasic?

Based on the above guidelines, we prepared eight RTP-PolyGILs

(Scheme 1, 2a–d and 3a–d) by the ion exchange between four

PolyGILs (Scheme 1, 1a–d) and two sulfonated phosphine ligands,

the monodentate TPPTS and the bidentate 2,7-bis(SO₃Na)Xantphos

50(SX). The PolyGILs had different capping groups (methoxyl or

phenoxyl) and/or different EO number (n = 4 or 16). The selected

ligands TPPTS¹³and SX¹⁴already have rich applications in biphasic

catalysis, especially for the hydroformylation of alkenes in ILs. The

ion exchange reaction gave a high yield of 85%–95% with <5%

55residual CH₃SO₃Na in the product according to ¹H NMR analysis.

80

Fig. 1 Solubility of RTP-PolyGILs in solvents of different polarity.

This journal is © The Royal Society of Chemistry 2019

Green Chemistry, 2019, 00, 1-3 | 3

Green Chemistry

Page 4 of 14

ARTICLE

Green Chemistry

Figure 1 illustrates the solubility of RTP-PolyGILs in various

These results suggested that the melting tem_Vp_iee_wra_At_ru_ticr_le_{e On}a_ln_ind_e

solvents of different polarity as well as their miscibility with 1-octene solubility of the RTP-PolyGILs could be easil^Dy^Ore^I:g¹u⁰l^.a¹⁰te³⁹d^/Cby^9Gc^Ch⁰a⁰n⁶g⁷in⁹g^F

(as the substrate of hydroformylation and hydrogenation) and n- 60the EO number and capping group of the polyether chain, and

nonanal (as the product of hydroformylation), using starting provided an important basis for constructing highly effective and

5

PolyGILs, TPPTS, and SX for comparison.

long life catalytic system.

For the water-soluble phosphine ligands TPPTS and SX, both were

Hydroformylation of Higher Olefins in a Homogeneous Catalysis-

insoluble in EtOH and i-PrOH, although TPPTS was slightly soluble Biphasic Separation (HCBS) System Using Rh-RTP-PolyGILs Catalysts.

while SX was easily soluble in MeOH at room temperature. Both were 65Considering the attractive properties of RTP-PolyGIL, we believed

insoluble in polar solvents (MeCN and THF), weakly polar solvents that its potential could be explored further by developing the highly

10(CHCl₃, CH₂Cl₂, t-BuOMe, Et₂O, and toluene), and apolar solvent (n- effective and long life catalytic system. Thus, we set off to devise a

heptane), and neither was soluble in 1-octene or n-nonanal. system using the RTP-PolyGILs that could harvest the advantages of

The PolyGILs were soluble in polar solvents (H₂O, MeOH, EtOH, i- both homogeneous and biphasic catalysis. We deployed this

PrOH, MeCN, and THF) and some weakly polar solvents (CHCl₃and 70ambitious plan by exploiting the phase behavior and miscibility of

CH₂Cl₂) but insoluble in some ether solvents (t-BuOMe and Et₂O) and RTP-PolyGILs, and evaluated this novel system using the

15in alkanes such as n-heptane, indicating that PolyGILs had some hydroformylation of higher olefins as the model reaction.

amphiphilicity thanks to the polyether chain. It is worth noting that

Since the RTP-PolyGILs were readily miscible with polar solvents

the solubility of the PolyGILs in toluene depended on the length of like alcohols (MeOH or EtOH) and weakly polar solvents like toluene

the polyether chain. For example, 1a ([CH₃(EO)₄TMG][OMs]) and 1c 75or CH₂Cl₂, but immiscible with some less polar or apolar solvents like

([Ph(EO)₄TMG][OMs]) were insoluble in toluene, both of which had 4 t-BuOMe and n-heptane, we naturally envisioned that these two sets

20EO units, and became the lower layer in a liquid/liquid biphasic of solvents could serve as the reaction (solvent I) and extraction

system with toluene. In contrast, 1b ([CH₃(EO)₁₆TMG][OMs]) and 1d (solvent II) medium for the hydroformylation reaction, respectively.

([Ph(EO)₁₆TMG][OMs]) were soluble in toluene, and both of them Specifically, a homogeneous catalysis-biphasic separation (HCBS)

had 16 EO units on the chain. Hence, the lipophilicity of the PolyGILs 80system that combined advantages of both homogeneous (high

grew as the polyether chain extended¹⁷.

25 Replacing the Na⁺ion in both TPPTS and SX by the polyether separation) could be concocted as follows (Figure 2).

guanidinium cation gave RTP-PolyGILs that inherited the solubility of In the first step A, the Rh precursor Rh(acac)(CO)₂, a catalytic

activity, and good selectivity) and biphasic catalysis (easy catalyst

the PolyGILs and showed the consistency with PolyGILs in solubility amount of RTP-PolyGILs (RTP-PolyGILs/Rh = 5:1–20:1 mol/mol,

in polar solvents (H₂O, MeOH, EtOH, i-PrOH, MeCN and THF), weak 85equivalent to 0.025–0.4 mol% of 1-alkenes), the 1-alkene, and the

polar solvents (CHCl₃, CH₂Cl₂, t-BuOMe, Et₂O and toluene) and apolar

30n-heptane. Compared with TPPTS and SX, RTP-PolyGILs seemed to

adapt better to solvents of different polarity and could be readily

miscible with H₂O, MeOH, EtOH, i-PrOH, MeCN, THF, CHCl₃, CH₂Cl₂,

or toluene. Hence, these solvents can be used as the medium for

forming the homogeneous catalytic system, and solvents such t-

35BuOMe, Et₂O, or n-heptane that are immiscible with RTP-PolyGILs

can be used to extract the reaction products.

Since the hydroformylation reaction converts 1-octene into n-

nonanal, the solubility of PolyGILs and RTP-PolyGILs in 1-octene and

n-nonanal was examined additionally at room temperature and at

40100 °C (hydroformylation reaction temperature). Both were

immiscible with the reactant 1-octene either at room temperature or

at 100 °C, while the situation became more complex in n-nonanal.

Only

1a

([CH₃(EO)₄TMG][OMs])

and

2b

([CH₃(EO)₁₆TMG]₃[P(C₆H₄SO₃)₃]) were miscible with n-nonanal at

45both room temperature and 100°C, presumably because the balance

between hydrophiles and lipophiles resembled that of n-nonanal.

Some RTP-PolyGILs, such as 1b ([CH₃(EO)₁₆TMG][OMs]), 2c

([Ph(EO)₄TMG]₃[P(C₆H₄SO₃)₃]), 2d ([Ph(EO)₁₆TMG]₃[P(C₆H₄SO₃)₃]),

and 3b ([CH₃(EO)₁₆TMG]₂[Xantphos-(SO₃)₂]), which were immiscible

50in n-nonanal at room temperature, became easily soluble in n-

nonanal at 100°C, showing the temperature-dependent solubility.

Most of RTP-PolyGILs (1c, 1d, 2a, 3a, 3c, and 3d) were immiscible in

n-nonanal neither at room temperature nor at 100 °C. The

immiscibility at room temperature could thus assist the

55hydroformylation reaction to ensure that the product aldehyde could

be separated naturally from the RTP-PolyGILs, thus inhibiting the

leaching of RTP-PolyGILs (or catalyst) into the product phase.

Fig. 2 Homogeneous catalysis-biphasic separation (HCBS) system for the highly

effective hydroformylation with Rh-RTP-PolyGILs.

90solvent I as the reaction medium would be mixed to form a

homogeneous system at reaction temperature (100°C). Note that no

solvent IL would have to be used. Then the hydroformylation would

proceed under pressured syngas. Here the reaction solvent I would

4 | Green Chemistry, 2019, 00, 1-3

This journal is © The Royal Society of Chemistry 2019

Page 5 of 14

Green Chemistry

ARTICLE

dissolve not only the Rh precursor and RTP-PolyGILs but also the

olefin substrate and product aldehydes, and most solvents in Figure

1 (alcohols, toluene, etc.) are suitable for this purpose. In the second

step B, upon completion, the system would be cooled to room

temperature, and after the release of syngas, the catalytic system

would remain homogeneous. The reaction solvent I would then be

removed under reduced pressure, and the extraction solvent II

would be introduced at <10°C. Note that the extraction solvent could

be obviated for the RTP-PolyGILs immiscible with aldehyde at room

View Article Online

Table 1. Hydroformylation of C₈–C₁₄1-alkenes in the HCBS system using Rh-

DOI: 10.1039/C9GC00679F

RTP-PolyGILs catalysts ^a

CO/H₂

Rh-2 or Rh-3

m

5

+

CHO

m

solvent

m = 5, 7, 9 and 11

CHO

d

Ligand or

Solvent ^b

(mL)

C ^c

S_ald

TOF ^f

(h⁻¹)

Entry

l/b ^e

RTP-PolyGIL

(%)

10temperature. At this point, a liquid/liquid or liquid/solid two phase

state would be formed for the mixture, with the organic phase in the

upper layer containing the product aldehydes, and the lower layer

retaining the Rh-RTP-PolyGILs (the third step C). Obviously, the

extraction solvent must dissolve the alkene and the aldehyde

15product but reject the RTP-PolyGILs, and alkanes such as n-heptane

would be readily suitable for this purpose. Next, the organic phase

containing the aldehyde could then be separated from Rh-RTP-

PolyGILs simply by decanting, and fresh olefin in the reaction solvent

I could be introduced to the recycled Rh-RTP-PolyGILs phase to start

20the next cycle (the fourth step D). Furthermore, for the

hydroformylation of lower olefins (C <6), the extraction step could be

replaced by distillation to separate the aldehyde and Rh-RTP-

PolyGILs.

Based on the principles envisioned above, we first evaluated the

25activity and selectivity of Rh-RTP-PolyGILs for the hydroformylation

of C₈–C₁₄1-alkenes in homogeneous systems, using PPh₃and TPPTS,

as well as Xanthene and SX as reference ligands. Toluene (PhMe) and

alcohols (MeOH and EtOH) were selected as the reaction solvent I,

representing less polar solvent and the polar solvent respectively.

30Alcohols (mainly MeOH and EtOH) were chosen because they are

relatively green solvents with low boiling point (easy recovery), and

for homogeneous reactions catalyzed by transition metals, often

they may participate in the catalytic cycle to give positive solvent

effects. A minimum amount of the reaction solvent, determined by

35cloud titration¹⁸, was added to form a homogeneous system.

Table 1 lists the experimental results. The catalytic activity was

evaluated by measuring the initial turnover frequency (TOF) within

the first 30 min or 2 h of the reaction. In toluene (entry 1, Table 1),

PPh₃gave an ultrahigh initial TOF to >25000 h⁻¹, good aldehyde

40selectivity (S_ald), and ordinary l/b ratio for the hydroformylation of

1-octene, which reflected the performance of a typical

homogeneous catalyst. Changing the reaction medium to MeOH

(entry 2, Table 1) afforded a modest increase in the rate, which

indicated a positive solvent effect and could be attributed to the

45facilitated catalytic cycle when MeOH served as a temporary ligand

to complex with the Rh center. In the case of TPPTS, which is unlike

PPh₃and insoluble in toluene and EtOH, the reaction activity (TOF)

declined drastically by 75%–80%, which was also accompanied by

poor aldehyde selectivity and low l/b ratio (entries 3 and 4, Table 1).

50Although substituting MeOH for EtOH improved the reaction rate

remarkably (entry 5, Table 1), mainly because TPPTS is slightly

soluble in MeOH, the TOF value was still far below the level of PPh₃

in toluene and MeOH. These results indicated that the low solubility

of ligands severely damaged the catalytic activity, so we expect

1

PPh₃

TPPTS

2b

PhMe 5.5

MeOH 5.5

PhMe 5.5

EtOH 5.5

MeOH 5.5

PhMe 20

MeOH 4

MeOH 5.5

MeOH 4

MeOH 7

MeOH 16

MeOH 23

EtOH 7

68.2

75.7

81.3

98.3

68.0

27.2

58.9

61.6

71.3

71.9

74.3

62.9

64.2

59.7

60.9

69.3

22.8

46.3

63.2

32.8

37.2

35.6

42.9

38.2

41.0

28.4

32.7

93.2

93.6

68.1

62.9

47.6

89.6

94.8

86.5

87.6

88.1

86.7

82.5

90.7

87.2

88.4

99.8

95.6

97.3

93.9

95.6

95.7

94.2

94.3

94.8

95.5

96.0

95.8

2.6

2.7

25432

28348

5537

2

3 ^g

4 ^g

5 ^h

6 ⁱ

7

1.5

1.8

6183

3.0

12947

4889

2.8

PPh₃

2a

2.6

22328

21303

24992

25323

25771

20767

21835

20814

21529

27667

2180

8

2.6

9

2b

2.6

10

2c

2.7

11

2d

2.6

12 ^j

13 ^k

14 ^l

15 ^m

16 ⁿ

17

2b

2.7

2b

2.3

2b

2.7

2b

EtOH 7

2.2

2b

MeOH 2

PhMe 4.5

MeOH 4.5

PhMe 10

MeOH 3

EtOH 3

0.3

Xanthene

SX

31.6

34.5

22.3

35.5

34.0

35.7

34.6

36.4

35.9

37.3

37.5

18

19 ^o

4509

3b

1484

20

3b

3138

21

3b

3558

22

3a

EtOH 3

3352

23

3c

EtOH 3

4045

24

3d

EtOH 3

3621

25 ^j

26 ^l

27 ^m

3b

EtOH 5

3918

3b

EtOH 6

2726

3b

EtOH 7

3132

^aReaction conditions: p (1:1 H₂/CO) = 7.0 MPa, T = 100 °C, Rh(acac)(CO)₂= 1.0

mg, 1-alkene/Rh = 20000, rate of stirring = 400 rpm; entries 1–16: ligand or

RTP-PolyGIL/Rh = 10, t = 0.5 h; entries 17–27: ligand or RTP-PolyGIL/Rh = 5, t =

2 h. ^bThe volume of solvents added was based on the mixability measured by

cloud titration according to literature¹⁸.

c

Conversion = (converted 1-

alkene)/(unconverted 1-alkene + converted 1-alkene) × 100%, the conversion

rate of 1-alkenes was determined on GC with the n-hexane (for 1-octene), n-

octane (for 1-decene), n-decane (for 1-dodecene), n-dodecane (for 1-

d

tetradecene), and cyclohexane (styrene) as internal standards. S_aldis the

aldehyde selectivity, S_ald= (linear acetal + 2-methylaldehyde)/(linear acetal + 2-

e

methylaldehyde + isomerized alkenes + acetals). l/b is the ratio of linear

f

aldehyde to branched aldehyde (2-methyl aldehyde). Turnover frequency

g

(TOF): mol (aldehyde) per mol (rhodium) per hour. TPPTS is insoluble in

toluene and EtOH, 1-octene/Rh = 5000. ^hTPPTS is slightly soluble in MeOH. ⁱ1-

octene/Rh = 10000. ^j1-decene/Rh = 20000. ^k1-dodecene/Rh = 5000, t = 8 min.

^l1-dodecene/Rh = 20000. ^m1-tetradecene/ Rh = 20000. ⁿstyrene/Rh = 20000.

^o1-octene/Rh = 5000.

55

good results from the RTP-PolyGILs because they could readily adapt

to various solvents and thus promote the catalytic activity. We first

This journal is © The Royal Society of Chemistry 2019

Green Chemistry, 2019, 00, 1-3 | 5

Green Chemistry

Page 6 of 14

ARTICLE

Green Chemistry

examined the performance of 2b in toluene. Surprisingly, although

2b was readily soluble in toluene, it exhibited similar activity to TPPTS

in toluene (entry 6 vs. entry 3, Table 1). We initially assumed that this

unexpectedly low rate might have been related to the high toluene

dosage that reduced the concentration of 1-octene, because that 20

mL toluene was needed to dissolve the olefin and 2b into a

homogeneous system according to the mixability test. To verify this

assumption, we increased the amount of toluene to 20 mL for PPh₃

(from entry 1 to entry 7, Table 1) and repeated the experiment, but

View Article Online

DOI: 10.1039/C9GC00679F

5

55

Fig. 3 Illustration of recycling of Rh-2a catalyst in the Rh-2a/1-octene/MeOH HCBS

hydroformylation system, a: corresponding to B in Figure 2; b and c: corresponding

to C in Figure 2; d: corresponding to D in Figure 2.

10found that the TOF dropped by only about 10%, suggesting that the

amount of toluene was not the main factor that impaired activity.

We next suspected the polyether structure of the RTP-PolyGIL 2b.

Although 2b could be dissolved in toluene, its hydrophilic polyether

chains could not be completely solvated by the less polar toluene

15solvent, and the polyether chains might curl into a micelle-like

structure to result in large steric hindrance around the Rh active

species and inhibit the activity of Rh-2b. The new assumption

motivated us to try highly polar alcohol solvents to improve the

results. As expected, 2a–d all exhibited much higher initial activity in

20MeOH than in toluene, with impressively high TOF ranging in 21000–

26000 h⁻¹, aldehyde selectivity of 86%–88%, and l/b of 2.6–2.7

(entries 8–11, Table 1). This performance is comparable to those of

PPh₃in toluene (entries 8–11 vs. entry 1, Table 1), and corresponds

to the catalytic behavior that a homogeneous-catalyst should have.

25The results could be attributed to the solvent effect of MeOH and the

hydrogen bonds that could be formed between the fully solvated

polyether chains and the hydroxyl groups of MeOH. In this way the

polyether chains became more spatially extended and no longer

impaired the activity. Note that the structure of the polyether

30guanidinium cation had no significant effect on the activity and

selectivity, because there was no limitation in mass transport. The

reaction rate was related only to the phosphine anion moiety, but

not to the polyether cation moiety, proving to be the kinetics of

homogeneous catalytic reaction. To evaluate if this system was

35general enough, the scope of substrates was studied next using

different alkenes in Rh-2b/MeOH. After we found that the amount

of MeOH required to form homogeneous systems increased with

rising chain length of the olefin, for example, 16 mL for 1-decene

(entry 12, Table 1) and 23 mL for 1-dodecene (entry 13, Table 1), we

40switched from MeOH to EtOH. The amount of EtOH needed was

successfully reduced to 7 mL for 1-dodecene and 1-tetradecene. In

the homogeneous systems with alcohol, the chain length of olefin

had little effect on the activity and selectivity. Compared with 1-

octene, only a modest drop in activity was observed for C₁₀–C₁₄

45olefins (entries 12–15 vs. entry 9, Table 1). As an exception, the

hydroformylation of styrene favored the branched aldehyde (l/b =

0.3), and gave a TOF of 27667 h⁻¹(entry 16, Table 1).

60(entry 17, Table 1). The sulfonated Xantphos (SX) was soluble in

MeOH and gave a higher TOF of 4500 h⁻¹and l/b ratio of 34.5 (entry

18, Table 1). Unsurprisingly, similar to 2b, the activity of 3b in toluene

dropped to 1484 h⁻¹(entry 19 vs. entry 17, Table 1) and bounced

back in alcohols to 3138 h⁻¹(MeOH) and 3558 h⁻¹(EtOH) (entries 20

65and 21 vs. entry 18, Table 1), close to the activity of SX in MeOH. Since

EtOH exhibited a more positive solvent effect than MeOH (entry 21

vs. entry 20, Table 1), it was selected as the solvent for subsequent

tests. The TOF value of 3300–4100 h⁻¹was also achieved for 3a, 3c

and 3d in EtOH (entries 22–24, Table 1), and the high activity was

70maintained when the substrates were extended to C₁₀–C₁₄1-alkenes

(entries 25–27, Table 1). All experiments of 3 in EtOH showed the

aldehyde selectivity of 94%–96% and l/b ratio of 34-37.5

(corresponding to the linear selectivity of 97%–98%). Using

sulfoxantphos anion to replace TPPTS anion successfully increased

75the l/b ratio from 2.2–2.7 to 34–37.5, which indicated that the

catalytic performance of Rh catalyst can be flexibly regulated by

adjusting the counterion structure of RTP-PolyGILs.

GC and GC/MS analyses indicated that the main by-products were

isomerized alkenes, and there was no hydrogenation product in all

80cases. Besides, only <1% acetals were formed when alcohols were

used as solvents.

To improve the mediocre selectivity of RTP-PolyGILs 2a–d for

linear aldehydes, we examined the diphosphine sulfoxantphos-

50functionalized polyether guanidinium ILs 3a–d (entries 17–27, Table

1) in the follow-up study, because it has been reported that the

Xantphos type ligands generally exhibit excellent linear selectivity in

hydroformylation^15,19. Firstly, Xantphos was used as a reference in

toluene to give a high l/b ratio of 31.6 with TOF being 2180 h⁻¹

Fig.

4 Recycling of Rh-2a catalyst in the Rh-2a/1-octene/MeOH HCBS

85hydroformylation system, S_l-ald= linear aldehyde/(linear aldehyde + 2-methyl

aldehyde), S_ace: the combined selectivity to linear acetal and branched acetal;

reaction conditions: p (1:1 H₂/CO) = 5.0 MPa, T = 100 °C, t = 1 h, Rh(acac)(CO)₂=

1.0 mg, 2a/Rh = 20, 1-octene/Rh = 5000, 1.0 μL of N,N,N’,N’- tetramethylguanidine

was added in cycles 1, 6, and 9.

6 | Green Chemistry, 2019, 00, 1-3

This journal is © The Royal Society of Chemistry 2019

Page 7 of 14

Green Chemistry

ARTICLE

NMR spectrum of the freshly prepared Rh-2a catalyst (Figure 5). In

View Article Online

45the ³¹P NMR spectrum of the freshly Rh-2a ^Dca^Ot^Ia^:l¹y⁰s^.t¹⁰(³a⁹,^/Cin^9GFi^Cg⁰u⁰re⁶⁷5⁹)^F,

the trivalent phosphine (δ = –3.90, or see Figure S 41) and a small

amount of phosphine oxide (δ = 28.07, or see Figure S 41) could be

clearly observed, and the two groups of double peaks at the δ 32.2 (J

= 128 Hz) and δ 38.8 (J = 138 Hz) respectively revealed the formation

50of both HRh(CO)(2d)₃and HRh(CO)₂(2d)₂as the precursors of the

active catalyst^9c. By contrast, the ³¹P NMR spectrum (b, in Figure 5)

of the recovered Rh-2a catalyst indicated that only the

resonance peak of the phosphine oxide could be observed, while the

signals of the trivalent phosphine and the precursors of active

55catalyst disappeared completely or might be too low to be observed,

which could be a evidence that the oxidation of 2a caused the

decomposition of the active catalyst precursors, in turn leading to

the decreased selectivity and the increased Rh loss (runs 9 and 10, in

Figure 4). One tenth of the initially loaded 2a was supplemented into

60the reaction in the eleventh cycle, which almost restored the

selectivity to the original level, further proving that ligand oxidation

was the main cause of increased Rh loss. A similar result was also

reported in a previous work by Lin et al²¹.

The formation of acetals was not negligible (up to 15% of the total

65products) in our initial recycling attempt, which severely reduced the

aldehyde yield. GC-MS analysis revealed the presence of a trace

amount of nonanoic acid in the products, which was responsible for

the formation of acetals. The acid was probably generated by the

oxidation of aldehydes and accumulated in the IL phase due to its

70insolubility in n-heptane. To solve this problem, we attempted to add

a trace amount of tetramethylguanidine (TMG) into the 2a phase to

neutralize, as TMG is readily soluble in RTP-PolyGILs. The addition of

TMG (runs 1, 6, and 9 in Figure 4) effectively inhibited the acetal

selectivity (S_ace) to < 5%.

Fig. 5 Comparison of ³¹P NMR spectra of freshly prepared and recovered Rh-2a

catalysts.

5

The high catalytic activity of the HCBS system prompted us to focus

next on the long term stability of Rh catalyst. We respectively used

the Rh-2a/1-octene/MeOH and Rh-3a/1-octene/EtOH HCBS systems

as models and concentrated on the separation, recovery, and

recycling of the Rh catalyst. Figure 3 illustrates the separation of Rh-

102a in the 1-octene/MeOH HCBS system, in which the individual

picture corresponds to the steps described in Figure 2. The system

was homogeneous after the reaction was terminated (a, in Figure 3)

and naturally split into two liquid phases at room temperature (b, in

Figure 3) after the removal of MeOH. The product aldehydes in the

15top layer were isolated, while Rh-2a adhered to the bottom and the

wall of the tube in the form of a jelly (b, in Figure 3). This

corresponded to the phase transition of 2a and its insolubility in n-

nonanal shown in the beginning. Next, n-heptane was added as an

extraction solvent, giving a colorless and transparent organic phase

20showing no significant Rh loss (c, in Figure 3). The upper organic

phase was easily decanted in air to recover Rh-2a (d, in Figure 3), and

fresh 1-octene and MeOH were then added to start the new cycle.

Encouragingly, Figure 4 shows that the recycled Rh-2a possessed

a long service life and allowed high olefin conversion (C > 95%), good

25aldehyde selectivity (S_ald), and almost a constant linear selectivity (S_l-

_ald) of 67%–72% corresponding to l/b = 2.1–2.5 in ten consecutive

cycles. Only the last two cycles exhibited a moderate decline in

aldehyde selectivity, along with an increase in the isomerization of

alkene. The total turnover number (TTON) reached 45232 in a total

30reaction time of only 10 hours, which is an unprecedented efficiency.

The ICP-AES analysis indicated that Rh leaching into the organic

phase in a single cycle was 0.04–0.4 ppm (0.03%–0.3% of the Rh

loaded) in the first eight runs but rose to 2–5 ppm (1.5%–3.5% of the

Rh loaded) in the last two cycles. Clearly, the rise in the isomerization

35after the eighth cycle could be associated with the corresponding

increase in Rh loss, attributable to the formation of oil-soluble

[HRh(CO)₄] species that readily catalyzes the isomerization of olefin

and easily dissolves in the organic phase²⁰. This sudden change might

have been caused by the oxidation of the phosphine moiety on 2a

40since our separation operation, which was carried out in air, had

already been carried out many times for the preceding cycles.

75 For more rigorously examinating the stability of the Rh-2a catalyst,

another series of recycling tests of the Rh-2a catalyst were

performed using the same reaction conditions as those shown in Fig.

4. The kinetic curves of hydroformylation were plotted by recording

80

Fig. 6 Long-term stability of Rh-2a catalyst in the Rh-2a/1-octene/MeOH HCBS

hydroformylation system, reaction conditions are the same as in Figure 4.

To verify this speculation, the recovered Rh-2a catalyst after the

tenth cycle was submitted to ³¹P NMR and compared with the ³¹P

This journal is © The Royal Society of Chemistry 2019

Green Chemistry, 2019, 00, 1-3 | 7

Green Chemistry

Page 8 of 14

ARTICLE

Green Chemistry

seven cycles. Only in the eighth cycle, the TOF value dropped below

View Article Online

10000 h^-1.

DOI: 10.1039/C9GC00679F

15 The high efficiency and good stability were also achieved in the

recycling experiment of the Rh-3a/EtOH HCBS system (Figure 7). The

conversion, aldehyde selectivity (S_ald) and linear selectivity (S_l-ald

)

were maintained above 93%, 94% and 97% (i.e., l/b > 32) over six

consecutive runs, respectively. The TTON value reached 26805, and

20each cycle restricted Rh leaching to 0.02–0.08 ppm (0.01%–0.06% of

the Rh loaded). The acetal selectivity (S_ace) in the products was kept

<3.5% by adding a trace amount of TMG.

This impressive long term stability of the Rh-catalyst, as we

contend, resulted from the structural design in integrating the ligand

25into the solvent IL and unique properties of RTP-PolyGIL. We showed

that the RTP-PolyGILs were completely soluble in MeOH and EtOH,

which greatly increased the reaction rate due to the positive solvent

effect of alcohols and shortened the reaction time to reduce the

probability of ligand oxidization. Even though the separation of Rh

30catalyst from the product aldehydes was carried out in air, the

negative effects of ligand oxidation were reflected in the reaction

only in the last two cycles (Figure 4). Besides, Rh leaching was

effectively inhibited during separation and recycling because 2a and

3a are immiscible with the product aldehydes. Moreover, the

35polyether guanidinium cations surrounding the rhodium active

center might act as a large steric barrier and an electrostatic shield

effectively to stabilize the Rh catalyst and prevent the formation of

Fig.

7 Recycling of Rh-3a catalyst in the Rh-3a/1-octene/EtOH HCBS

hydroformylation system, S_l-ald= linear aldehyde/(linear aldehyde + 2-methyl

aldehyde), S_ace: the combined selectivity to linear acetal and branched acetal;

5reaction conditions: p (1:1 H₂/CO) = 5.0 MPa, T = 100 °C, t = 6 h, Rh(acac)(CO)₂=

1.0 mg, 3a/Rh = 20, 1-octene/Rh = 5000, 1.0 μL N,N,N’,N’-tetramethylguanidine

was added in cycles 2, 4, and 6.

the pressure drop of the syngas during the catalyst runing (Fig. 6).

10The linearity of the graphs, the good parallelism of TOF values (Run

1-7: TOF = 19400–24200 h^-1) and the almost constant l/b ratio of 2.1–

2.3 all confirmed that the Rh-2a catalyst remained stable in the first

inactive rhodium clusters^8g,20,22

.

Table 2. Comparison of the HCBC system with some classic systems for Rh-catalyzed hydroformylation of 1-alkenes

Solvent

Solvent IL or

TOF

(h⁻¹)

Rh Loss ^l

Entry

System

IL/alkene

(mol %)

Ligand

l/b

TTON

Ref.

reaction medium

(ppm or %)

j

1 ^a

2 ^b

Biphasic

[BMIM][BF₄]

[BMIM][PF₆]

44.6 ^j

n.r. ^k

[BDMIM]₃[P(C₆H₄SO₃)₃]

Cationic Xantphos

1380

2.2

31248

3500

0.3–4.1

<0.07%

9c

8c

52

517

239

21.3

3 ^c

4 ^d

scCO₂-Biphasic

[OctMIM][NTf₂]

—

[PMIM][Ph₂P(C₆H₄SO₃)]

[OctMIM][Ph₂P(C₆H₄SO₃)]

3.0

—

<0.1

4b

a stable state mixture of

olefin and aldehyde

[BMIM][BF₄]

3.2

5–10

10a

j

5 ^e

6 ^f

SILC

0.6 ^j

9.3 ^j

—

[BMIM]₃[P(C₆H₄SO₃)₃]

3900

298

2.4

n.r.

2.8

n.r.

2.1

5a

23

j

Thermoregulated [CH₃(EO)₁₆NEt₃][MeSO₃]

TPPTS

11950

2979

45232

0.45%–0.82%

0.4–1.9

7 ^g

8 ^h

CO₂-Switching

Our system

None

P[m-PhN=C(Me)(NMe₂)]₃

11046

21303

24

—

2a

2.6

0.04–0.4,

(0.03%–0.3%)

0.02–0.08,

9 ⁱ

Our system

None

—

3a

3352

35.7

26805

—

(0.01%–0.06%)

^aBiphasic IL system: p (1:1 H₂/CO) = 4.0 MPa, T = 100 °C, Rh(acac)(CO)₂as Rh precursor, ligand/Rh = 10, 1-hexene as substrate. ^bBiphasic IL system: p (1:1 H₂/CO) =

3.0 MPa, T = 100 °C, Rh(acac)(CO)₂as Rh precursor, ligand/Rh = 2, 1-octene as substrate, the ligand is a Xantphos ligand with two guanidinium tags. ^cSupercritical

CO₂-IL biphasic system (continuous flow): total pressure (CO₂/H₂/CO) = 20 MPa, T = 100 °C, Rh(acac)(CO)₂as Rh precursor, 1-octene as substrate. ^dSupercritical CO₂

biphasic system (continuous flow): Rh catalyst dissolved in substrate–product mixture, total pressure (CO₂/H₂/CO) = 14 MPa, T = 100 °C, Rh(acac)(CO)₂as Rh

precursor, ligand/Rh = 15, 1-octene as substrate. ^eSupported ionic liquid catalysis: p (1:1 H₂/CO) = 10 MPa, T = 100 °C, Rh(acac)(CO)₂as Rh precursor, ligand/Rh = 10,

f

1-hexene as substrate. Thermoregulated phase separation system: toluene and n-heptane as solvents, p (1:1 H₂/CO) = 5.0 MPa, T = 110 °C, RhCl₃·3H₂O as Rh

precursor, ligand/Rh = 8, 1-dodecene as substrate. ^gCO₂-Switching system: toluene and H₂O as solvents, p (1:1 H₂/CO) = 2.0 MPa, T = 100 °C, Rh(acac)(CO)₂as Rh

precursor, ligand/Rh = 50, 1-octene as substrate. ^hHCBS system: MeOH as solvent, p (1:1 H₂/CO) = 7.0 MPa, T = 100 °C, Rh(acac)(CO)₂as Rh precursor, ligand/Rh =

10, 1-octene as substrate. ⁱHCBS system: ligand/Rh = 5, other conditions the same as in entry 8. ^jCalculated based on data from ref. ^kn.r.: not reported. ^lRh leaching

is expressed as a concentration (ppm) of Rh in the product aldehydes or a percentage of lost Rh to the Rh loaded.

8 | Green Chemistry, 2019, 00, 1-3

This journal is © The Royal Society of Chemistry 2019

Page 9 of 14

Green Chemistry

ARTICLE

View Article Online

DOI: 10.1039/C9GC00679F

Table 2 compares our system with various classic catalytic

systems, including biphasic IL system, supercritical CO₂biphasic

system, supported ionic liquid catalysis (SILC) system,

thermoregulated phase separation system, and CO₂-switching

system. Note that P-FILs were used in the first four systems (entries

1-4, Table 2). However, unlike our RTP-PolyGILs, these P-FILs only as

ionic ligands did not demonstrate the solvent properties of the room

temperature IL since there were no polyether chains on their onium

Table 3. Hydrogenation of C₈–C₁₄1-alkenes catalyzed by Rh-2b in the

HCBS system ^a

H₂

5

Rh-2b

m

solvent

m = 5, 7, 9 and 11

d

Solvent ^b

(mL)

Conv. ^c

S_alk

TOF

(h^-1)

Entry

Alkene

(%)

10ions. So they can only be used for the catalytical reaction in the

presence of the solvent IL or using a mixture of the substrate and the

product as the reaction medium. Specifically, in biphasic IL systems

using P-FILs (entries 1 and 2, Table 2)^{8c, 9c}, a large amount of solvent

IL is required, which limits the mass transport and reduces the

15catalyst TOF. The continuous flow reactions in the scCO₂biphasic

system (entries 3 and 4, Table 2)^4b,10aalso suffer from mass transport

limitation and require high total pressure (up to 14–20 MPa)

although the steady state mixture of the substrate and the product

has replaced the solvent IL to support catalyst^10a. The SILC system

20(entry 5, Table 2)^5asucceeds in cutting down the use of solvent IL and

improving mass transport, but it still lags far behind homogeneous

system in terms of TOF. Thermoregulated phase separation and the

CO₂-switching system (entries 6 and 7, Table 2)^23,24overcome the

mass transport problem and have the characteristics of

25homogeneous catalysis, but they require the use of weakly polar or

apolar solvents (e.g., toluene and n-heptane) for eventual phase

separation, which to some extent restricted the catalytic activity

because the reaction must be run in the weakly polar organic phase.

From these comparisons, our HCBS system (entries 8 and 9, Table 2)

30is far superior in all aspects, including the amount of IL use, catalytic

activity, linear selectivity, long term catalyst stability, and Rh

leaching. Thus far, the current results remain unparalleled by any

existing literature reports. Undoubtedly the integration of both

phosphine ligand and solvent IL gave RTP-PolyGILs that radically

35improved the efficiency, economy, and environmental friendliness in

utilizing the multi-functional IL at only catalytic amount.

1 ^e

2

1-Octene

MeOH 2

EtOH 2

99.0

96.0

98.8

78.3

99.6

99.2

88.2

85.0

96.3

90.5

88.5

88.7

81.4

68.5

46.2

79.9

31.6

50.2

3580

3400

3505

2547

2727

1831

2820

1073

1935

1-Octene

3

1-Octene

4

1-Decene

MeOH 3.5

EtOH 2

5

1-Decene

6

1-Dodecene

1-Tetradecene

MeOH 6

EtOH 2

7

8

MeOH 13

9

EtOH 2

^aReaction conditions: p (H₂) = 6.0 MPa, T = 80°C, t = 15 min, RhCl₃·3H₂O

b

1.0 mg, 2b/Rh = 10, alkenes/Rh = 1,000, the rate of stirring = 400 rpm.

The volume of solvents added was based on mixability determinations

that were carried out by cloud titration method according to literature¹⁸.

c

Conversion of 1-alkenes was determined on GC with the n-hexane (for

1-octene), n-octane (for 1-decene), n-decane (for 1-dodecene) and n-

d

dodecane (for 1- tetradecene) as internal standards. Selectivity to

alkanes. ^eLigand = PPh₃.

the alkenes increased, the reactivity declined and more isomerized

alkenes could be observed (entries 4–9, Table 3). Using EtOH gave a

stronger positive solvent effect than using MeOH (entry 4 vs. 5, 6 vs.

607, 8 vs. 9, Table 3).

There were two ways to separate the Rh-catalyst from product

alkane. Vacuum distillation was used for n-octane and n-decane

because their boiling point was lower, whereas n-dodecane and n-

tetradecane whose boiling point was higher separated automatically

65from Rh-2b after removal of the alcohol owing to 2b being immiscible

with alkanes. The Rh-2b catalyst recycled from the 1-octene/MeOH

HCBS system showed a long service life in twelve

Hydrogenation of Higher Olefins in an HCBS System Using Rh-

RTP-PolyGIL Catalyst. The successful application of RTP-PolyGILs in

the hydroformylation of 1-alkenes encouraged us to further

40investigate their universality in other transition metal catalytic

reactions, the first of which came to mind was catalytic

hydrogenation. Obviously, based on our work above, the

hydrogenation of alkenes would be the most suitable model

reaction.

45 In Table 3, the activity and selectivity of Rh-2b in the hydro-

genation of C₈-C₁₀1-alkenes under the HCBS system were evaluated

by measuring the initial TOF within the first 15 min of the reaction,

and contrasted with PPh₃ligand. A catalytic amount of 2b,

equivalent to 1 mol% of alkene substrate, was used in the HCBS

50system. For the hydrogenation of 1-octene in MeOH and EtOH, Rh-

2b gave a activity of 3400–3500 h^-1comparable to those of PPh₃in

MeOH (entries 2–3 vs. entry 1, Table 3), and the only by-products

were isomerized alkenes. The amount of MeOH required to form

homogeneous systems increased as the chain length of the alkene

55substrates increased, which is consistent with the trend shown in

Table 1, so EtOH is used instead of MeOH. When the chain length of

70Fig.

8 Recycling of Rh-2b catalyst in the Rh-2b/1-octene/MeOH HCBS

hydrogenation system, reaction conditions: p (H₂) = 6.0 MPa, T = 80 °C, t = 5 h,

RhCl₃·3H₂O = 1.0 mg, 2b/Rh = 20, 1-octene/Rh = 1000.

Green Chemistry, 2019, 00, 1-3 | 9

Green Chemistry

Page 10 of 14

ARTICLE

Green Chemistry

used, and the chromatographic conditions were that a 80 ˚C oven

55temperature was kept for 2 min followed by an increase at a rate of

continuous cycles, giving >98% average conversion, >96% average

alkane selectivity (S_alk), <4% average isomerized alkenes selectivity

(S_iso), and >11000 TTON (Figure 8), demonstrating the good

universality for catalytic reactions.

We believe the concept of ligand–IL integration can be further

applied to other catalytic reactions, and an application in asymmetric

hydrogenation is currently underway. One of our preliminary

patents²⁵involving this paper had been filed in as early as 2013 and

authorized in 2016.

View Article Online

DOI: 10.1039/C9GC00679F

6 ˚C/min up to 220 ˚C, and held for 20 min. For hydrogenation, a DB–

35 capillary columns (FID detector) was used, and the

chromatographic conditions were that a 35 ˚C oven temperature was

kept for 5 min followed by an increase at a rate of 8 ˚C/min up to 200

60˚C, and held for 20 min. GC–MS was used to qualitatively analyze the

products of the hydroformylation and hydrogenation reactions on an

Agilent 6890/5973 GC-MS apparatus using a DB-35MS capillary

column. Melting point (T_m) was measured by DSC using a Mettler

5

1

Toledo DSC1 instrument under N₂atmosphere. H NMR, ¹³C NMR

65and ³¹P NMR (85% phosphoric acid as the internal standard) spectra

were recorded on a Bruker AVANCE 500 or an AVANCE III 600

spectrometer. Mass spectroscopy was measured on a Q–ToF Ultima

Global mass spectrometer. ICP–AES analysis was performed on an

IRIS Intrepid II XSP apparatus to measure Rh leaching.

10Conclusions

In conclusion, based on the concept of ligand–IL integration, we

have synthesized a class of room temperature phosphine-

functionalized polyether guanidinium ionic liquids (RTP-

PolyGILs) that exist as liquids at room temperature. The

15incorporation of polyether chains converted the sulfonated

phosphine ligands from high melting solids into room

temperature ionic liquids (ILs), allowing the complete fusion of

phosphine ligands and solvent ILs both in structure and

properties. The melting temperature and solubility of the RTP-

20PolyGILs could be easily regulated by changing the EO number

and capping group of the polyether chain. Based on the unique

properties of RTP-PolyGILs, we established a homogeneous

catalysis−biphasic separation (HCBS) system for Rh-catalyzed

hydroformylation of higher olefins using only a catalytic amount

25of RTP-PolyGILs. In our HCBS system, the advantages of both

homogeneous (high activity, and good selectivity) and biphasic

catalysis (easy catalyst separation) were combined, as well as

the catalytic activity and linear selectivity could be flexibly

controlled by changing the type of the sulfonated ligand anion

30on RTP-PolyGILs. The best catalytic activity reached 25000 h⁻¹,

and the l/b ratio could be increased from 2.2:1 to 37.5:1 by

changing the counterion from TPPTS to sulfoxantphos.

Meanwhile, the Rh catalyst exhibited a long-term stability,

giving TTON up to 45000 h⁻¹with Rh leaching of 0.02–0.4 ppm.

35The RTP-PolyGILs were additionally applied to the catalytic

hydrogenation reaction of olefins, which demonstrated their

universality in catalytic reactions. Therefore, the efficient,

economical, and green use of IL was realized in our HCBS

system, thanks to the integration of both ligand and IL.

70 Preparation of [CH₃(EO)₄TMG]₃[P(C₆H₄SO₃)₃] (2a) and its

characterization. Under Ar atmosphere, TPPTS·4H₂O (1.5 g, 2.3

mmol), 1a [CH₃(EO)₄TMG][OMs] (2.7 g, 6.7 mmol) and acetonitrile

(25 mL) were added to a 100 mL Schlenk flask, and the mixture was

stirred at room temperature for 72 h. The reaction was terminated

75when ¹H NMR analysis indicated that residual CH₃SO₃Na in the

products was <5%. The mixture was filtered and then acetonitrile

was evaporated under reduced pressure to give 2a as a yellow

1

viscous liquid. Yield: 2.9 g (86%). Characterizations: H NMR (500.0

MHz, D₂O): δ = 7.79 (d, 3H, P-Ph-H), 7.74 (d, 3H, P-Ph-H), 7.51 (t, 3H,

80P-Ph-H), 7.45 (t, 3H, P-Ph-H), 3.63–3.53 (m, 53H, OCH₂CH₂), 3.31 (s,

9H, OCH₃), 2.87 (s, 36H, 6×N(CH₃)₂); ¹³C NMR (125.7 MHz, CDCl₃): δ =

162.58, 146.95, 136.55, 134.51, 131.02, 128.36, 126.83, 71.82, 70.43,

70.29, 69.48, 58.92, 44.69, 39.79; ³¹P NMR (202.4 MHz, D₂O): δ = -

5.33; HRMS (Q−ToF MS, ES+): m/z = 262.2122, calcd. for C₁₂H₂₈O₃N₃

85([CH₃(OCH₂CH₂)₃TMG]⁺): 262.2125; m/z = 306.2382, calcd. for

C₁₄H₃₂O₄N₃([CH₃(OCH₂CH₂)₄TMG]⁺): 306.2387; m/z = 350.2642,

calcd. for C₁₆H₃₆O₅N₃([CH₃(OCH₂CH₂)₅TMG]⁺): 350.2649; m/z =

394.2903, calcd. for C₁₈H₄₀O₆N₃([CH₃(OCH₂CH₂)₆TMG]⁺): 394.2912;

HRMS (Q−ToF MS, ES−): m/z = 166.3133 (z = 3), calcd. for

90C₁₈H₁₂O₉PS₃: 166.3133; m/z = 249.9729 (z = 2), calcd. for C₁₈H₁₃O₉PS₃:

249.9735.

Preparation of [CH₃(EO)₁₆TMG]₃[P(C₆H₄SO₃)₃] (2b) and its

characterization: Under Ar atmosphere, TPPTS·4H₂O (0.5 g, 0.78

mmol), 1b [CH₃(EO)₁₆TMG][OMs] (2.1 g, 2.3 mmol) and acetonitrile

95(25 mL) were added to a 100 mL Schlenk flask, and the mixture was

stirred at room temperature for 72 h. The reaction was terminated

when ¹H NMR analysis indicated that residual CH₃SO₃Na in the

products was <5%. The mixture was filtered and then acetonitrile

was evaporated under reduced pressure to give 2b as a yellow

100viscous liquid at room temperature. As the temperature fell below

40Experimental

Materials and methods. The 1-alkenes, Rh(acac)(CO)₂, RhCl₃·3H₂O,

and TPPTS ligand were purchased from Acros Chimica Co.. The

Xantphos ligand was obtained from Bei Jing Greenchem Technology 10 °C, 2b solidified gradually. Yield: 2.1 g (90%). Characterization: ¹H

Co. in China. The 2,7-bis(SO₃Na) Xantphos ligand¹⁵and PolyGILs¹²

45were synthesized according to literature methods. Regular Schlenk

techniques were applied in all experimental operations with Ar as the

inert atmosphere. All reagents and solvents were purified and

processed according to standard methods²⁶and deoxygenated

strictly before use. The hydroformylation and hydrogenation

50reactions were carried out in a 60 mL stainless steel (316 L)

autoclave. GC analysis was used to determine the conversion and

selectivity of the hydroformylation and hydrogenation reactions. For

hydroformylation, a OV–101 capillary column (FID detector) was

NMR (500.0 MHz, D₂O): δ = 7.87 (d, 3H, P-Ph-H), 7.82 (d, 3H, P-Ph-H),

7.59 (t, 3H, P-Ph-H), 7.52 (t, 3H, P-Ph-H), 3.84–3.44 (m, 200H,

OCH₂CH₂), 3.39 (s, 9H, OCH₃), 2.95 (s, 36H, 6×N(CH₃)₂); ¹³C NMR

105(125.7 MHz, CDCl₃): δ = 162.07, 146.54, 136.03, 134.00, 130.39,

127.85, 126.27, 71.31, 69.93, 69.63, 68.97, 58.40, 44.13, 39.31; ³¹P

NMR (202.4 MHz, D₂O): δ = −5.11; HRMS (Q−ToF MS, ES+): m/z =

746.4994, calcd. for C₃₄H₇₂O₁₄N₃([CH₃(OCH₂CH₂)₁₄TMG]⁺): 746.5009;

m/z = 790.5256, calcd. for C₃₆H₇₆O₁₅N₃([CH₃(OCH₂CH₂)₁₅TMG]⁺):

110790.5271; m/z = 834.5518, calcd. for C₃₈H₈₀O₁₆N₃([CH₃(OCH₂CH₂)₁₆

10 | Green Chemistry, 2019, 00, 1-3

Page 11 of 14

Green Chemistry

ARTICLE

TMG]⁺): 834.5533; m/z

= 878.5779, calcd. for C₄₀H₈₄O₁₇N₃mixture was stirred at room temperature for 72h. The reaction was

View Article Online

1

([CH₃(OCH₂CH₂)₁₇TMG]⁺): 878.5795; m/z = 922.6041, calcd. for terminated when H NMR analysis indicated that residual CH₃SO₃Na

DOI: 10.1039/C9GC00679F

C₄₂H₈₈O₁₈N₃([CH₃(OCH₂CH₂)₁₈TMG]⁺): 922.6057; HRMS (Q−ToF MS, 60in products was <5%. The mixture was filtered and then acetonitrile

ES−): m/z = 166.3132 (z = 3), calcd. for C₁₈H₁₂O₉PS₃: 166.3133; m/z = was evaporated under reduced pressure to give 3a as a yellow

1

5

249.9730 (z = 2), calcd. for C₁₈H₁₃O₉PS₃: 249.9735.

viscous liquid. Yield: 1.5 g (87%). Characterization: H NMR (500.0

Preparation of [Ph(EO)₄TMG]₃[P(C₆H₄SO₃)₃] (2c) and its MHz, D₂O): δ = 8.01 (s, 2H), 7.03–6.92 (m, 22H), 3.71–3.39 (m, 32H,

characterization. Under Ar atmosphere, TPPTS·4H₂O (1.5 g, 2.3 OCH₂CH₂), 3.29 (s, 6H, OCH₃), 3.24 (m, 4H, OCH₂CH₂), 2.83 (s, 24H),

mmol), 1c [Ph(EO)₄TMG][OMs] (3.1 g, 6.7 mmol) and acetonitrile (25 651.64 (s, 6H); ¹³C NMR (125.7 MHz, CD₃OD): δ = 163.02, 154.36,

mL) were added to a 100 mL Schlenk flask, and the mixture was 141.13, 136.95, 134.67, 131.43, 130.35, 129.93, 129.53, 127.65,

10stirred at room temperature for 72 h. The reaction was terminated 125.63, 72.12, 70.68, 70.52, 70.12, 59.06, 45.23, 40.15, 35.83, 31.65;

when ¹H NMR analysis indicated that residual CH₃SO₃Na in the ³¹P NMR (202.4 MHz, D₂O): δ = −16.89; HRMS (Q−ToF MS, ES+): m/z

products was <5%. The mixture was filtered and then acetonitrile = 262.2120, calcd. for C₁₂H₂₈O₃N₃([CH₃(OCH₂CH₂)₃TMG]⁺): 262.2125;

was evaporated under reduced pressure to give 2c as a yellow 70m/z = 306.2380, calcd. for C₁₄H₃₂O₄N₃([CH₃(OCH₂CH₂)₄TMG]⁺):

1

viscous liquid. Yield: 3.2 g (90%). Characterization: H NMR (500.0 306.2387; m/z = 350.2640, calcd. for C₁₆H₃₆O₅N₃([CH₃(OCH₂CH₂)₅

15MHz, D₂O): δ = 7.85 (d, 3H, P-Ph-H), 7.81 (d, 3H, P-Ph-H), 7.56 (t, 3H, TMG]⁺): 350.2649; m/z

= 394.2901, calcd. for C₁₈H₄₀O₆N₃

P-Ph-H), 7.46 (t, 3H, P-Ph-H), 7.38 (t, 6H, O-Ph-H), 7.03 (m, 9H, O-Ph- ([CH₃(OCH₂CH₂)₆TMG]⁺): 394.2912; HRMS (Q−ToF MS, ES−): m/z =

H), 4.20–3.30 (m, 70H, OCH₂CH₂), 2.87 (s, 36H, 6×N(CH₃)₂); ¹³C NMR 368.0450 (z = 2), calcd. for C₃₉H₃₀O₇P₂S₂: 368.0460; m/z = 737.0978

(150.9 MHz, CDCl₃): δ = 162.39, 158.45, 146.62, 136.41, 134.54, 75(z = 1), calcd. for C₃₉H₃₁O₇P₂S₂: 737.0992.

130.76, 129.30, 128.34, 126.71, 120.74, 114.36, 70.54–69.98, 69.52,

Preparation of [CH₃(EO)₁₆TMG]₂[Xantphos-(SO₃)₂] (3b) and its

2069.33, 67.11, 44.57, 39.63; ³¹P NMR (202.4 MHz, D₂O): δ = −5.32; characterization. Under Ar atmosphere, 2,7- bis(SO₃Na)Xantphos

HRMS (Q−ToF MS, ES+): m/z = 324.2274, calcd. for C₁₇H₃₀O₃N₃(1.0 g, 1.3 mmol), 1b [CH₃(EO)₁₆TMG][OMs] (2.3 g, 2.5 mmol) and

([Ph(OCH₂CH₂)₃TMG]⁺): 324.2282; m/z = 368. 2535, calcd. for acetonitrile (25 mL) were added to a 100 mL Schlenk flask, and the

C₁₉H₃₄O₄N₃([Ph(OCH₂CH₂)₄TMG]⁺): 368.2544; m/z = 412.2795, calcd. 80mixture was stirred at room temperature for 72h. The reaction was

for C₂₁H₃₈O₅N₃([Ph(OCH₂CH₂)₅TMG]⁺): 412.2806; m/z = 456.3054, terminated when ¹H NMR analysis indicated that residual CH₃SO₃Na

25calcd. for C₂₃H₄₂O₆N₃([Ph(OCH₂CH₂)₆TMG]⁺): 456. 3068; m/z = in products was <5%. The mixture was filtered and then acetonitrile

500.3312, calcd. for C₂₅H₄₆O₇N₃([Ph(OCH₂CH₂)₇TMG]⁺): 500.3330; was evaporated under reduced pressure to give 3b as a yellow

HRMS (Q−ToF MS, ES−): m/z = 166.3133 (z = 3), calcd. for viscous liquid at room temperature. As the temperature fell below

C₁₈H₁₂O₉PS₃: 166.3133; m/z = 249.9729 (z = 2), calcd. for C₁₈H₁₃O₉PS₃: 8510 °C, 3b solidified gradually. Yield: 2.6 g (86%). Characterization: ¹H

249.9735.

NMR (500.0 MHz, D₂O): δ = 8.01 (s, 2H), 7.02 (m, 22H), 3.72–3.66 (m,

30 Preparation of [Ph(EO)₁₆TMG]₃[P(C₆H₄SO₃)₃] (2d) and its 131H, OCH₂CH₂), 3.37 (s, 6H, OCH₃), 3.33 (m, 4H, OCH₂CH₂), 2.91 (s,

characterization. Under Ar atmosphere, TPPTS·4H₂O (1.5 g, 2.3 24H), 1.65 (s, 6H); ¹³C NMR (125.7 MHz, CD₃OD): δ = 163.08, 154.30,

mmol), 1d [Ph(EO)₁₆TMG][OMs] (6.6 g, 6.7 mmol) and acetonitrile 141.11, 136.94, 134.66, 131.46, 130.32, 130.05, 129.61, 127.51,

(25 mL) were added to a 100 mL Schlenk flask, and the mixture was 90125.62, 72.09, 70.67, 70.14, 59.05, 45.20, 40.15, 35.81, 31.67; ³¹P

stirred at room temperature for 72 h. The reaction was terminated NMR (202.4 MHz, D₂O): δ = −17.16; HRMS (Q−ToF MS, ES+): m/z =

35when ¹H NMR analysis indicated that residual CH₃SO₃Na in the 746.5000, calcd. for C₃₄H₇₂O₁₄N₃([CH₃(OCH₂CH₂)₁₄TMG]⁺): 746.5009;

products was <5%. The mixture was filtered and then acetonitrile m/z = 790.5262, calcd. for C₃₆H₇₆O₁₅N₃([CH₃(OCH₂CH₂)₁₅TMG]⁺):

was evaporated under reduced pressure to give 2d as a yellow 790.5271;

m/z

=

834.5524,

calcd.

for

C₃₈H₈₀O₁₆N₃

viscous liquid at room temperature. As the temperature fell below 95([CH₃(OCH₂CH₂)₁₆TMG]⁺): 834.5533; m/z = 878.5785, calcd. for

10 °C, 2d solidified gradually. Yield: 6.7 g (95%). Characterization: ¹H C₄₀H₈₄O₁₇N₃([CH₃(OCH₂CH₂)₁₇TMG]⁺): 878.5795; m/z = 922.6047,

40NMR (500.0 MHz, D₂O): δ = 7.87 (d, 3H, P-Ph-H), 7.82 (d, 3H, P-Ph-H), calcd. for C₄₂H₈₈O₁₈N₃([CH₃(OCH₂CH₂)₁₈TMG]⁺): 922.6057; HRMS

7.59 (t, 3H, P-Ph-H), 7.51 (t, 3H, P-Ph-H), 7.41 (t, 6H, O-Ph-H), 7.10– (Q−ToF MS, ES−): m/z = 368.0453 (z = 2), calcd. for C₃₉H₃₀O₇P₂S₂:

7.05 (m, 9H, O-Ph-H), 4.24–3.39 (m, 200H, OCH₂CH₂), 2.96 (s, 36H, 368.0460; m/z = 737.0982 (z = 1), calcd. for C₃₉H₃₁O₇P₂S₂: 737.0992.

6×N(CH₃)₂); ¹³C NMR (150.9 MHz, CDCl₃): δ = 162.53, 158.59, 146.81, 100 Preparation of [Ph(EO)₄TMG]₂[Xantphos-(SO₃)₂] (3c) and its

136.52, 134.43, 130.10, 129.24, 128.36, 126.77, 120.66, 114.44, characterization. Under Ar atmosphere, 2,7- bis(SO₃Na)Xantphos

4570.34, 69.58, 67.13, 44.61, 39.74; ³¹P NMR (202.4 MHz, D₂O): δ = (1.0 g, 1.3 mmol), 1c [Ph(EO)₄TMG][OMs] (1.1 g, 2.4 mmol) and

−5.32; HRMS (Q−ToF MS, ES+): m/z

= 808.5156, calcd. for acetonitrile (25 mL) were added to a 100 mL Schlenk flask, and the

C₃₉H₇₄O₁₄N₃([Ph(OCH₂CH₂)₁₄TMG]⁺): 808.5165; m/z = 852.5419, mixture was stirred at room temperature for 72h. The reaction was

calcd. for C₄₁H₇₈O₁₅N₃([Ph(OCH₂CH₂)₁₅TMG]⁺): 852.5427; m/z = 105terminated when ¹H NMR analysis indicated that residual CH₃SO₃Na

896.5681, calcd. for C₄₃H₈₂O₁₆N₃([Ph(OCH₂CH₂)₁₆TMG]⁺): 896.5690; in products was <5%. The mixture was filtered and then acetonitrile

50m/z = 940.5942, calcd. for C₄₅H₈₆O₁₇N₃([Ph(OCH₂CH₂)₁₇TMG]⁺): was evaporated under reduced pressure to give 3c as a yellow

1

940.5952; HRMS (Q−ToF MS, ES−): m/z = 166.3134 (z = 3), calcd. for viscous liquid. Yield: 1.6 g (89%). Characterization: H NMR (500.0

C₁₈H₁₂O₉PS₃: 166.3133; m/z = 249.9732 (z = 2), calcd. for C₁₈H₁₃O₉PS₃: MHz, D₂O): δ = 8.01 (s, 2H), 7.07–6.61(m, 32H), 4.20–2.90 (m, 40H,

249.9735.

110OCH₂CH₂), 2.66 (s, 24H), 1.57 (s, 6H); ¹³C NMR (125.7MHz, CD₃OD): δ

Preparation of [CH₃(EO)₄TMG]₂[Xantphos-(SO₃)₂] (3a) and its = 163.97, 160.30, 154.52, 142.33, 137.94, 135.12, 131.02, 130.59,

55characterization. Under Ar atmosphere, 2,7- bis(SO₃Na)Xantphos 129.82, 129.53, 127.96, 125.89, 122.02, 115.71, 71.55, 71.33, 70.92,

(1.0 g, 1.3 mmol), 1a [CH₃(EO)₄TMG][OMs] (1.0 g, 2.5 mmol) and 70.51, 68.62, 45.93, 40.22, 36.09, 31.92; ³¹P NMR (202.4 MHz, D₂O):

acetonitrile (25 mL) were added to a 100 mL Schlenk flask, and the δ = −17.08; HRMS (Q−ToF MS, ES+): m/z = 324.2274, calcd. for

Green Chemistry, 2019, 00, 1-3 | 11

Green Chemistry

Page 12 of 14

ARTICLE

Green Chemistry

C₁₇H₃₀O₃N₃([Ph(OCH₂CH₂)₃TMG]⁺): 324.2282; m/z = 368.2535, calcd. respectively, with intense stirring. After 15 min,_Vt_ieh_we_Arr_tie_cla_ec_Ot_ni_lo_inn_e

for C₁₉H₃₄O₄N₃([Ph(OCH₂CH₂)₄TMG]⁺): 368.2544; m/z = 412.2795, was terminated by placing the autoclave in an ice bath. Upon

DOI: 10.1039/C9GC00679F

calcd. for C₂₁H₃₈O₅N₃([Ph(OCH₂CH₂)₅TMG]⁺): 412.2806; m/z = 60releasing the gas, the solution was submitted to GC analysis.

456.3053, calcd. for C₂₃H₄₂O₆N₃([Ph(OCH₂CH₂)₆TMG]⁺): 456.3068;

5

m/z = 500.3312, calcd. for C₂₅H₄₆O₇N₃([Ph(OCH₂CH₂)₇TMG]⁺):

Conflicts of interest

There are no conflicts of interest to declare.

500.3330; HRMS (Q−ToF MS, ES−): m/z = 368.0453 (z = 2), calcd. for

C₃₉H₃₀O₇P₂S₂: 368.0460; m/z

= 737.0977 (z = 1), calcd. for

C₃₉H₃₁O₇P₂S₂: 737.0992.

Preparation of [Ph(EO)₁₆TMG]₂[Xantphos-(SO₃)₂] (3d) and its

Acknowledgements

10characterization. Under Ar atmosphere, 2,7- bis(SO₃Na)Xantphos

(1.0 g, 1.3 mmol), 1d [Ph(EO)₁₆TMG][OMs] (2.5 g, 2.5 mmol) and

acetonitrile (25 mL) were added to a 100 mL Schlenk flask, and the

mixture was stirred at room temperature for 72h. The reaction was

terminated when ¹H NMR analysis indicated that residual CH₃SO₃Na

15in products was <5%. The mixture was filtered and then acetonitrile

was evaporated under reduced pressure to give 3d as a yellow solid

We gratefully acknowledge the financial support from the

65National Natural Science Foundation of China (No.21576144),

the Natural Science Foundation of Shandong Province (No.

ZR2014BM009), Taishan Scholars Projects of Shandong (No.

ts201511033), and the National Natural Science Foundation of

China (No.21878164).

1

at room temperature. Yield: 2.7 g (85%). Characterization: H NMR

(500.0 MHz, D₂O): δ = 7.99 (s, 2H), 7.15–6.83(m, 32H), 3.99–3.23 (m,

124H, OCH₂CH₂), 2.82 (s, 24H), 1.61 (s, 6H); ¹³C NMR (150.9 MHz,

20CD₃OD): δ = 163.97, 160.30, 154.50, 142.30, 137.91, 135.14, 131.00,

130.56, 129.87, 129.57, 127.98, 125.88, 121.93, 115.69, 74.70-71.32,

70.86, 70.53, 68.52, 45.93, 40.25, 36.10, 31.93; ³¹P NMR (202.4 MHz,

D₂O): δ = −17.46; HRMS (Q−ToF MS, ES+): m/z = 808.5155, calcd. for

C₃₉H₇₄O₁₄N₃([Ph(OCH₂CH₂)₁₄TMG]⁺): 808.5165; m/z = 852.5418,

25calcd. for C₄₁H₇₈O₁₅N₃([Ph(OCH₂CH₂)₁₅TMG]⁺): 852.5427; m/z =

896.5680, calcd. for C₄₃H₈₂O₁₆N₃([Ph(OCH₂CH₂)₁₆TMG]⁺): 896.5690;

m/z = 940.5940, calcd. for C₄₅H₈₆O₁₇N₃([Ph(OCH₂CH₂)₁₇TMG]⁺):

940.5952; HRMS (Q−ToF MS, ES−): m/z = 368.0452 (z = 2), calcd. for

70Notes and references

1

(a) P. Wasserscheid, T. Welton, Ionic Liquids in Synthesis, Wiley–VCH,

Weinheim, Germany, 2003; (b) P. J. Dyson and T. J. Geldbach, Metal

catalysed Reactions in Ionic Liquids, Springer, Dordrecht, The Netherlands,

2005 ; (c) C. Hardacre and V. Parvulescu, Catalysis in Ionic Liquids: From

Catalyst Synthesis to Application, The Royal Society of Chemistry,

Cambridge, UK, 2014; (d) T. Welton, Chem. Rev., 1999, 99, 2071. (e) P.

Wasserscheid and W. Keim, Angew. Chem. Int. Ed., 2000, 39, 3772; (f) R.

Sheldon, Chem. Commun., 2001, 2399; (g) J. Dupont, R. F. de Souza and P.

A. Z. Suarez, Chem. Rev., 2002, 102, 3667; (h) T. Welton, Coord. Chem. Rev.,

2004, 248, 2459; (i) V. I. Pârvulescu and C. Hardacre, Chem. Rev., 2007,

107, 2615; (j) M. Haumann and A. Riisager, Chem. Rev., 2008, 108, 1474;

(k) N. V. Plechkova and K. R. Seddon, Chem. Soc. Rev., 2008, 37, 123; (l) H.

Olivier-Bourbigou, L. Magna and D. Morvan, Applied Catalysis A: General,

2010, 373, 1; (m) J. P. Hallett and T. Welton, Chem. Rev., 2011, 111, 3508;

(n) F. Jutz, J. Andanson and A. Baiker, Chem. Rev., 2011, 111, 322; (o) S.

Zhang, J. Sun, X. Zhang, J. Xin, Q. Miao and J. Wang, Chem. Soc. Rev., 2014,

43, 7838; (p) T. L. Greaves and C. J. Drummond, Chem. Rev., 2015, 115,

11379; (q) C. Dai, J. Zhang, C. Huang and Z. Lei, Chem. Rev., 2017, 117, 6929;

(r) R. L. Vekariya, Journal of Molecular Liquids, 2017, 227, 44.

75

C₃₉H₃₀O₇P₂S₂: 368.0460; m/z

30C₃₉H₃₁O₇P₂S₂: 737.0992.

= 737.0980 (z = 1), calcd. for

80

General procedure for evaluating the catalyst activity and

selectivity of hydroformylation of C₈-C₁₄1-alkenes in the HCBS

system using Rh-RTP-PolyGILs catalysts. Under Ar atmosphere, a 60

mL autoclave was loaded with Rh(acac)(CO)₂(1.0 mg, 3.88  10^3

35mmol), 10 equivalents of 2 (3.88  10^2mmol, 55 mg for 2a, 116 mg

for 2b, 62 mg for 2c, 124 mg for 2d) or 5 equivalents of 3 (1.94  10^2

mmol, 26 mg for 3a, 47 mg for 3b, 28 mg for 3c, 49 mg for 3d), and

solvent (MeOH or EtOH, see Table 1 for the volumes). Subsequently,

2  10⁴equivalents of C₈–C₁₄1-alkene (77.6 mmol, 12.0 mL for 1-

40octene, 14.5 mL for 1-decene, 17.0 mL for 1-dodecene, 19.5 mL for

1-tetradecene) was added along with the internal standard (see

Table 1), and the reaction temperature and syngas pressure were

brought to 100 °C and 7.0 MPa, respectively, with intense stirring.

After 30 min (for 2) or 2 h (for 3), the reaction was terminated by

45placing the autoclave in an ice bath. Upon releasing the syngas, the

solution was submitted to GC analysis.

General procedure for evaluating the catalyst activity and

selectivity of hydrogenation of C₈-C₁₄1-alkenes in the HCBS

system using Rh-2b catalyst. Under Ar atmosphere, a 60 mL

50autoclave was loaded with RhCl₃·3H₂O (1.0 mg, 3.8  10^3

mmol), 10 equivalents of 2b (114 mg, 3.8  10^2mmol), and

solvent (MeOH or EtOH, see Table 3 for the volumes).

Subsequently, 1  10³equivalents of C₈–C₁₄1-alkene (3.8 mmol,

0.6 mL for 1-octene, 0.7 mL for 1-decene, 0.85 mL for 1-

55dodecene, 1.0 mL for 1-tetradecene) was added along with the

internal standard (see Table 3), and the reaction temperature

and H₂pressure were brought to 80 °C and 6.0 MPa,

85

90

2

3

R. A. Sheldon, Green Chem., 2017, 19, 18.

(a) A. Riisager, R. Fehrmann, M. Haumann and P. Wasserscheid, Top.

Catal., 2006, 40, 91; (b) C. Van Doorslaer, J. Wahlen, P. Mertens, K.

Binnemans and D. De Vos, Dalton Trans ., 2010, 39, 8377; (c) C. Chiappe

and M. Malvaldi, Phys. Chem. Chem. Phys., 2010, 12, 11191.

95

4

5

(a) G. Smith, D. J. Cole-Hamilton, A. C. Gregory and N. G. Gooden,

Polyhedron, 1982, 1, 97; (b) P. B. Webb, M. F. Sellin, T. E. Kunene, S.

Williamson, A. M. Z. Slawin and D. J. Cole-Hamilton, J. Am. Chem. Soc.,

2003, 125, 15577.

(a) C. P. Mehnert, R. A. Cook, N. C. Dispenziere and M. Afeworki, J. Am.

Chem. Soc., 2002, 124, 12932; (b) H. Li, P. S. Bhadury, B. Song and S. Yang,

RSC Advances, 2012, 2, 12525; (c) T. Selvam, A. Machoke and W.

Schwieger, Applied Catalysis A: General, 2012, 445−446, 92; (d) B. Xin and

J. Hao, Chem. Soc. Rev., 2014, 43, 7171; (e) E. García-Verdugo, B. Altava,

M. I. Burguete, P. Lozanob and S. V. Luis, Green Chem., 2015, 17, 2693; (f)

B. Karimi, M. Tavakolian, M. Akbari and F. Mansouri, ChemCatChem, 2018,

10 (15), 3173.

100

105

12 | Green Chemistry, 2019, 00, 1-3

Page 13 of 14

Green Chemistry

ARTICLE

6

7

8

(a) R. Ballini, Eco-Friendly Synthesis of Fine Chemicals, The Royal Society

of Chemistry, Cambridge, UK, 2009; (b) J. H. Davis, Chem. Lett., 2004, 33,

7, 373; (c) C. P. Mehner, R. A. Cook, N. C. Dispenziere and E. J. Mozeleski,

View Article Online

DOI: 10.1039/C9GC00679F

Polyhedron, 2004, 23, 2679.

1072; (c) S. Lee, Chem. Commun., 2006, 1049; (d) M. Pucheault and M. 10 (a) P. B. Webb and D. J. Cole-Hamilton, Chem. Commun. 2004, 612; (b) A.

Vaultier, Top. Curr. Chem., 2009, 290, 83; (e) R. Giernoth, Angew. Chem.

Int. Ed., 2010, 49, 2834; (f) A. D. Sawant, D. G. Raut, N. B. Darvatkar and 40

C. Frisch, P. B. Webb, G. Zhao, M. J. Muldoon, P. J. Pogorzelec and D. J.

5

10

15

20

25

30

35

Cole-Hamilton, Dalton Trans. 2007, 5531.

M. M. Salunkhe, Green Chem. Lett. Rev., 2011, 4, 41; (g) C. Yue, D. Fang, L. 11 (a) S. Tang, G. A. Baker and H. Zhao, Chem. Soc. Rev., 2012, 41, 4030; (b)

Liu and T. Feng, J. Mol. Liq., 2011, 163, 99.

M. M. Cecchini, C. Charnay, F. De Angelis, F. Lamaty, J. Martinez and E.

Colacino, ChemSusChem, 2014, 7, 45; (c) Y. Qiao, W. Ma, N. Theyssen, C.

Chen and Z. Hou, Chem. Rev., 2017, 117, 6881.

(a) R. Šebesta, Enantioselective Homogeneous Supported Catalysis, The

Royal Society of Chemistry, Cambridge, UK, 2012; (b) R. Šebesta, I.

Kmentova and Š. Toma, Green Chem., 2008, 10, 484; (c) C. Huo and T. H. 4512 X. Jin, D. Yang, X. Xu and Z. Yang, Chem. Commun., 2012, 48, 9017.

Chan, Chem. Soc. Rev., 2010, 39, 2977; (d) B. Ni and A. D. Headley, Chem. 13 Y. Chauvin, L. Mussmann and H. Olivier, Angew. Chem. Int. Ed. Engl., 1995,

Eur. J., 2010, 16, 4426; (e) M. Lombardo and C. Trombini, ChemCatChem,

34 (23/24), 2698.

2010, 2, 135; (f) J. Limberger, B. C. Leal, A. L. Monteiro and J. Dupont, 14 J. Dupont, S. M. Silva and R. F. de Souza, Catal. Lett., 2001, 77 (1−3), 131.

Chem. Sci., 2015, 6, 77.

15 M. S. Goedheijt, P. C. J. Kamer and P. W.N.M. van Leeuwen, J. Mol. Cat. A:

Chem., 1998, 134, 243.

(a) C. C. Brasse, U. Englert, A. Salzer, H. Waffenschmidt and P. 50

Wasserscheid, Organometallics, 2000, 19, 3818; (b) F. Favre, H. Olivier- 16 Z. B. Zhou, H. Matsumoto and K. Tatsumi, Chem. Eur. J., 2005, 11, 752.

Bourbigou, D. Commereuc and L. Saussine, Chem. Commun., 2001, 1360; 17 J. Leicunaite, I. Klimenkovs, J. Kviesis, D. Zacs and J. P. Kreišmanis, C. R.

(c) P. Wasserscheid, H. Waffenschmidt, P. Machnitzki, K. W. Kottsieper and

Chim., 2010, 13, 1335.

O. Stelzer, Chem. Commun., 2001, 451; (d) R. P. J. Bronger, S. M. Silva, P. 18 A. Behr, G. Henze, D. Obst and B. Turkowski, Green Chem., 2005, 7, 645.

C. J. Kamer and P. W. N. M. van Leeuwen, Chem. Commun., 2002, 3044; 5519 M. Kranenburg, Y. E. M. Vanderburgt, P. C. J. Kamer, P. W. N. M. van

(e) D. J. Brauer, K. W. Kottsieper, C. Liek, O. Stelzer, H. Waffenschmidt and

P. Wasserscheid, J. Organomet. Chem., 2001, 630, 177; (f) K. W. 20 P. W. N. M. Van Leeuwen and C. Claver, Rhodium Catalyzed

Kottsieper, O. Stelzer and P. Wasserscheid, J. Mol. Catal. A: Chem., 2001, Hydroformylation, Kluwer Academic, Dordrecht, The Netherlands, 2002.

Leeuwen, K. Goubitz and J. Fraanje, Organometallics, 1995, 14, 3081.

175, 285; (g) R. P. J. Bronger, S. M. Silva, P. C. J. Kamera and P. W. N. M. 21 Q. Lin, W. Jiang, H. Fu, H. Chen and X. Li, Appl. Catal. A, 2007, 328, 83.

van Leeuwen, Dalton Trans., 2004, 1590; (h) T. E. Kunene, P. B. Webb and ⁶⁰22 A. Castellanos-Páez, S. Castillón, C. Claver, P. W. N. M. van Leeuwen and

D. J. Cole-Hamilton, Green Chem., 2011, 13, 1476; (i) K. L. Luska, K. Z.

Demmans, S. A. Stratton and A. Moores, Dalton Trans ., 2012, 41, 13533; 23 B. Tan, J. Jiang, Y. Wang, L. Wei, D. Chen and Z. Jin, Appl. Organomet.

(j) X. Jin, K. Zhao, F. Kong, F. Cui and D. Yang, Catal. Lett., 2013, 143, 839; Chem., 2008, 22, 620.

W. G. J. Lange, Organometallics, 1998, 17, 2543.

(k) X. Jin, K. Zhao, F. Cui, F. Kong, and Q. Liu, Green Chem., 2013, 15, 3236; 24 S. L. Desset and D. J. Cole-Hamilton, Angew. Chem. Int. Ed., 2009, 48, 1472.

(l) Y. Li, H. Liu, P. Wang, D. Yang, X. Zhao and Y. Liu, Faraday Discuss., 2016, ⁶⁵25 X. Jin, S. Li, K. Zhao, CN Patent 2013103701386, 2013.

190, 219; (m) Y. Li, P. Wang, H. Liu, Y. Lu, X. Zhao and Y. Liu, Green Chem., 26 W. L. F. Armarego and C. L. L. Chai, Purification of Laboratory Chemicals,

2016, 18, 1798; (n) C. S. Consorti, G. L. P. Aydos, G. Ebeling and J. Dupont,

Org. Lett., 2008, 10, 237.

5nd Edition, Elsevier/Butterworth-Heinemann, Burlington, MA, USA,

2003.

9

(a) M. F. Sellin, P. B. Webb and D. J. Cole-Hamilton, Chem. Commun., 2001,

781; (b) P. B. Webb, T. E. Kunene, D. J. Cole-Hamilton, Green Chem., 2005,

Green Chemistry, 2019, 00, 1-3 | 13

Green Chemistry

Page 14 of 14

View Article Online

DOI: 10.1039/C9GC00679F

Integration of Phosphine Ligand and Ionic Liquid Both in Structure and Property: An

Efficient and Economical Catalytic System for Homogeneous-Catalyst Recycling

• A table of contents entry:

* Colour graphic: maximum size 8 cm x 4 cm

* Text: one sentence, of maximum 20 words, highlighting the novelty of the

work

Article Doi

Integration of phosphine ligands and ionic liquids both in structure and properties-a new strategy for separation, recovery, and recycling of homogeneous catalyst

DOI: 10.1039/c9gc00679f

Source and publish data:

Authors:

Article abstract of DOI:10.1039/c9gc00679f

Full text of DOI:10.1039/c9gc00679f

Products guided by the article

R&D Labs maybe for 10486-19-8

Relevant to this article

Hot Product