CA1088442A - Fructose production by enzymatic isomerization of glucose - Google Patents

Abstract

IMPROVED FRUCTOSE PRODUCTION

ABSTRACT OF THE INVENTION
Fructose productivity and isomerase activity in immobilized beds or column operations employing isomerases obtained from Bacillus organisms are significantly improved by isomerizing a high solids feed syrup at pH
7.0-7.5 and 55°C to 60°C. Without adding cobalt to the feed streams con-tinuous column operation in excess of 4,000 hours and yielding greater than 3,500 pounds of a 42% fructose syrup for each pound of isomerase can be achieved.

Description

`.~
34~Z

BACKGROUND OF THE INVENTION
Fructose obtained by enzymatically isomerizing dextrose to fructose is extensively used by the food industry as a sucrose replace-ment.
~ .

Substantial fructose production costs are encountered by the need to fre~uently replace spent or deactivated glucose isomerases.
Enhanced fructose productivity by the glucose isomerase ls a desirable goal. Extensive research efforts have been expended towards obtaining maximum fructose productivity with the lowest possible amount of glucose - -~
isomerase. Many researchers have deemed the solution to the problem as being simply a matter of discovering a stable glucose isomerase. In this endeavor, the art has screened, mutated and prepared a multitude of di~er-ent glucose isomerase preparations. In testing the efficacy and suscept-ibility of these glucose isomerases to deactivation, the art has come to the realization that isomerases derived from different microbe sources - :; ' possess different enzymatic characteristics. Optimum isomerization condi-tions such as pH and temperature, isomerase activators (e.g., metal ion activators such as Co~+, manganese, etc.) and other processlng variables will depend uponparticular glucose isomerase type. In general, a greater disparity in isomerization conditions occur~ between isomerases derived from a different genera.
., ,: . :, -.
:
1 Immobilized isomerases are more stable than isomerases in a .--,~ water-soluble or unbound form. In general, immobilized glucose isomerases `:
l~ are better suited for commercial operations since they may be continuously .:
used in batch or continuous operations until exhausted. Upon exhaustion, --~
these isomerases are replaced with fresh immobilized isomerase. Most ~ . , .
~ ''.,:
~:, . . .
,

-2-,.

'. . , .. . .. , , , ; ; , ~ .

~6~44~

lsomerization reactions are conducted at temperatures and pHs which optimize the rate at which the isomerase converts dextrose to fructose.
Similar to o~her enæymes, isomerases are usually most stable against inactivation (including heat inactivation) and pos~less a higher enzymatic activity when used at their optimum isomerization pF[. Glucose isomerases presently used in the commercial production of fructose containing ~yrups characteristically exhibit improved stability and activity when the isomer-ization process is conducted in the presence of Co~ ions. Cobaltous salts are occasionally added to feed syrups for this purpose. It would be desirable to achieve a higher productivity without necessitating cobaltous ions.
, . ' ,:
A suggested glucose isomerization processlng modification ls to increase isomerization reaction temperature as isomerase activity decreases.
~n increase in the reactor temperature will accelerate the rate of fructose production as well as the rate at which the isomerase deactivates. The net effect is to lower the total fructose yield produced by the isomerase.

Isomerases reportedly produced by organisms belonging to the Bacillus genera include Bacillus stearothermophilus ATCC 31265, NR~L
B-3680, NRRL B-3681 and NRRL B-3682; Bacillus sp. NRRL B-5350 and NRRL
B-5351; Bacillus megaterium ATCC 15450; Bacillus fructosus ATCC 15451, 35c. (e.g., see U. S. Patent Nos. 3,826,714 by Suekane et al. and

3,306,752 by K. Ueda, West German printed Patent Application 2,164,342 .
filed under S.N. P 2,164.342.5 on December 23, 1971 by K. Aunstrup et al., Agri. Biol. Chem., Vol. 31~ No. 3, pages 284-292, 1967 by Danno et al.).
Heat or chemical treatment of viable cells containing intracellular isomerase, encapsulation, complexing with natural and synthetic poly-mers, immobilization within a binder matrix and numerous other means lV~3~344'~

for immobilizing isomerases have been suggested. Exemplary methods for immobilizing isomerases and enhancing enzymatic stability are disclosed in J. Appl. Chem. Biotechnol. 1974, 24, 663-676 by Rent et al. Agri. ;;
Biol. Chem., Vol. 30, ~o. 10, pages 1015-1023, 1966 by S. Yoshimura et al.
(e.g., see Holland Patent Application 73/12525 filed September 11, 1973 and assigned to ~ovo Terapeutisk Laboratorium, Brit:Lsh Patent Specifica-tion 1,274,158, U. S. Patent Nos. 3,821,082 by Lamm et al., 3,779,869 by Zienty, 3,694,314 by Lloyd et al., 3,788,945 by Thompson et al., and Brltish Patent Specification 1,356,283 by Monsanto Co., etc.).
-- ' ' ' .~,'".,"
A recent article entitled "Sweetzyme - A New Immobilized Glucose Isomerase," die Starke 27, Jahrg 1975/Nr. 7, pages 236-241 disclo~es immobilized isomerases of a Bacillus coaLulan origin. This article defines ' gluco~e isomerase productivity as the combined efect of activity and sta-¦ bility. At the more neutral pHs, cobalt ls deemed essential for fructo~e . .. .
productivity. In order to achleve optimum fructose syrup productivity in the absence of cobalt ions, the authors conclude that it is necessary to -ll conduct the continuous isomerization reaction at a relatively high alkaline i pH. For a continuous operation ~e.g., column isomerization) involving a short contact and reaction time between isomerase and syrup, optimum productivity in the ab~ence of Co~ is reportedly achieved at a pH above 8.0 (e.g., 8.1-8.5), a 40-45% solids level and 65C.

OBJECTS
! : ~ It is an ob~ect to prolong the useful llfe and fructose produc-tivity of immobilized glucose isomerases of a Bacillus origi~.

..
i 25 Another ob~ect is to improve the processing efficacy of isomeriz-ing glucose to fructose in column type reactors.

) , ~ ~,:':,.

_4_ ~18~44Z ~:
.
A still further object is to contlnuously isomerize a dextrose syrup to a fructose syrup under processing conditions which alleviates the formation of undesirable by-products.

THE INVENTION ~ - :
S According to the present invention, there i8 provided a process for : :
improving fructose syrup productivity by isomerizing dextrose to fructose ::
within an immobilized glucose isomerase bed in which the isomerase i5 characterized as being obtained from the Bacillus genera and exhibiting an :. :~ enhanced rate of isomerizing dextrose to fructose when the glucose isomexiza-; 10 tion reaction is conducted: (a) in the presence of Co~ ions and (b~ at a ~f temperature greater than about 60C., said process comprising:
~A) providing a refined dextrose syrup which is essentially free from Co+~ ions and containing on a total dry solids weight basis at least 90% monosaccharide, ~;~

(B) isomerizing the dextrose syrup to a fructose syrup by I passing the dextrose syrup through a bed of immobi-lized glucose isomerase maintained at a temperature of -` no more than about 60C. and a p~ between 7.0 and 7.5, .
', '.':
(C) recovering the fructose syrup while replenishing the ¦ 20 bed with fresh dextrose syrup, and ,: i ~ (D) continuing the isomerization of the dextrose syrup in `
1. said bed until the total isomerase bed activity has .
', been reduced to a value of less than 20% of its origi-l nal activity thus achieving improved fructose syrup ¦ 25 production from said bed.
f .. ..
- .:
..
; ,' -5- ~:

344~ ~
The present process employs a refined dextrose syrup having a high monosaccharlde solids content and essentially free from Co+~ ions. The present invention may be used in a process in which a reactor produces the desired fructose syrup product in a single pass or recycling the syrup through a reactor at a higher flow rate until the desired interconversion is achieved or with a plurality of reactors connected in series wherein the fructose con-tent is incrementally increased as the syrup flows through each reac~or in the series.
,'-:
Productivity of the isomerization process is generally enhanced by employing feed syrups of a high monosaccharide content. High dextrose con-version syrups containing more than 90% dextrose tdry solids weight basis -d.s.b.) or 95~ dextrose or higher ~especially at about 97 to 99~) are par-i ticularly useful eed syrups (e.g., see U. S. Patent Nos. 3,783~100 by R.
Larson et al. and 3,897,305 by T. ~urst).

The isomerized syrup quality and glucose isomerase bed productivity are adversely affected by inorganic and organic, non-saccharide syrup con-i taminants. Certain metal ions, in trace quantities, such as aluminum, copper, tin, zinc, mercury, calcium, etc. and anions which inactivate the isomerase , and/or react to orm insolubles, can reduce bed productivity. Such con-¦ 20 taminants may be removed rom the dextrose eed syrup by conventional cation i and anion resin treatment. Incomplete ion exchange treatment can also lead ~ to the development of insoluble floc or precipitates in the isomerization ¦ systemcausiDg pressure drops ln the isomerization column reactor which reduces -~
`i bed productivity. These ionic impurities may be sui~ably removed from the high dextrose feed syrup by single or double cation and anion exchange treat-ment (e.g., strong cation-wea~ anion-strong cation-weak anion).
~1 - , ' :' .
"

., 1~8~a~4Z
Organic substances typically present in high dextrose conversion syrups such as proteinaceous materials, color (e.g., HMF) and flavor con-taminants, etc. have a deleterious effect upon bed productivity and fructose syrup quality. Conventional ion exchange treatment cannot remove all of these undesirable organic substances from the feed syrup. Such organic sub-stances may be conveniently removed from the syrup by conventional means such as granular carbon, activated carbon treatment (e.g., at levels of -~
about 0.5 to about 2.0 parts by weight activated carbon for each 100 parts by weight dry syrup solids). Insoluble organic and inorganic substances are also desirably removed by conventional means from the feed syrup prior to the isomerization reaction. Refining the syrup by the sequential steps of removing the insolubles, carbon treatment and ion exchange treatment is generally satisactory.

The present isomerization reaction is conducted a temperatures, pH levels and other operating conditions which are conducive to microbial . . .
~ growth. Without adequate safeguards against microbial growth, the isomeriza- ;
: . .
tion reactors upon prolonged usage can become infested. This will adversely affect isomerase productivity and syrup quality. By ad~usting the feed syrup dry solids to more than 45% by weight, and preferably at least 50%
by weight, the microbial infestation problem is more easily controlled. Feed syrups containing more than 60% dry solids are generally too viscous for effective passage through the isomerization reactor. A feed syrup adjusted to a dry solids level ranging from about 50% to about 5~% by weight will , ~generally provide adequate flow rates through the reactor while minimizing mlcrobial infestation. Periodic or continuous treatment of the feed syrups with conventional bactericides can be used as a processing aid to reduce microblal lnfestation.

~` ' ~ ,.
,~: , :
.

l -7-4Z :: ~
."::, .: ' '.
In the present process, dextrose is isomerized to fructose by pass-ing a dextrose syrup through an immobilized bed containing an isomerase derived from an organism of the Bacillus genera. Suitable immobilized beds forconducting the isomerization reaction -lnclude conventional techniques - -and apparatus for confining the immobilized isomerase within an isomeriza~
tion reaction zone while permitting the passage of syrup through the bed.
Column type reactor systems operated in a single pass or recycling or a 1 '. .. ~'. '.
plurality of reactors connected in series may be utilized to convert the syrup to the desired fructose containing syrup product (e.g., fructose to -dextrose weight ratio between about 2:3 to about 1:1). Preferably, the isomerization process is conducted by continuously permitting a dextrose syrup to flow through a column reactor contalning immobilized glucose isomerase wlth the amount of glucose isomerase and flow rate of syrup through the column being ~ufficlent to lncrea8e the fructose syrup content to a level between ~;
~ 15 ~bout 44 to about 47% (monosaccharlde welght basis).

`I The isomerases employed in this invention are characterized as exhibiting an optimum isomerase activity within the range of 8.0-8.5 as dete~mined under standard assay conditions with an assay substrate of 60 :, ,, "
grams water, 40 grams anhydrous dextrose, 0.02M magnesium sulfate, 0.0035M
cobalt chlorlde at 65C. for one hour. Under these assay conditions, the lsomerase wlll produce more fructose between pH 8.0-8.5 than wlll be pro-i duced outside this range. If the above standard assay temperature is !~
`~ reduced (e.g., 60C. or less), these isomerases (under theseassay condi-l ~ tions) produce less fructose than they will at 65~C. or higher. Another . ,.
characteristic of the isomerase is that when cobalt chloride is excluded from the standard assay substrate and the assay is conducted at pH 7.5 and 60C. for one hour, the fructose production will be less than that produced when 0.0035M cobalt chloride is included as an assay ingredient.
'1 ' , '~ '. ' .

l -8 ~, . .. . .

,.. , . . , .. . .. , , .. : ,....... . :, ., -:: :;: . :. .. : .: .:: ~ . ,... .:

Although the isomerization process generally applies ;
to immobilized isomerases which are derived from the Bacillus ~
genera, the invention is particularly adapted for use with ~ ~ ;
isomerases obtained from the Bacillus coagulan family (e.g., ~ -NRRL B-5305 and NRRL B-5351) and immobilizecl in accordance with U.K. patent specification No. 1,444,539. These isomer-ases are more stable against inactivation when used in an iso-merization process with a stabilizing amount of Co + ion (e.g., between 0.0015-0.004M), possess a pH optimum of about 8.5 and optimum isomerization temperature well above 60C.

. . .
In practicing the invention, column type reactors may be suitably loaded with isomerase typically having an ac-tivity (per the standard assay conditions mentioned above) `
greater than about ~00 IGIU/gram of enzyme. The bed advan-ta~eously contains greater than about 3 x 106 IGI~ and pre-erably more than about 6 x 106 IGIU/per cubic oot of bed volume. When it is desired to produce about 45~ fructose ~ -~: .
(monosaccharide basis) in a single pass, column loading so as to permit an initial syrup flow rate through a fresh bed of about 0.05 to about 0.2 gallon/min. (usually at about 0.1) for each cubic foot of bed volume is generally suitable. The flow rate is proportionally reduced to compensate for deacti-vation of the bed as the isomerization process proceeds.

The enzymatic isomerization is conducted at a pH
between 7.0 to less than about 7.5. If the pH drops below the 7.0 level, the isomerase (without the presence of a sta-bilizing amount of Co++) is susceptible to permanent deact-ivation. At a pH in excess of 7.5, the total fructose pro-ductivity of the bed will similarly decrease because of iso-merase deactivation. An ancillary advantage of operating the isomerization process within this more neutral pH range is that it prevents formation of undesirable flavor . ~ . .
B -g- ~

.. . . . .. . .. . .... . . . . . . . .. . . . .. . . . . . .. .

and color bodies. In comparison to a pH 8.5 and 65C. process, `
fructose productivity of the bed is enhanced under the present process from about 5 to 10 fold. This enhanced fructose pro- `~
ductivity substantially reduces the overall total isomerase requirements to produce a given amount of fructose, and enables one to operate the reactor for a longer period of time without ~ ;
interrupting its operation to reload~ Similar to other enzymes, isomerases are typically more sensitive to deactivation when -used at a pH substantially below their optimum pH. Contrary ~
to expectations, the total bed fructose productivity is signi- ~ -ficantly increased by conducting the isomerization reaction without a stabilizing amount of cobalt (e.g., more than O.OOlM) at a pH level which is well outside its optimum isomerization pH range.

During the isomerization process, the dextrose rich syrup migrates into the immobilized isomerase particles. Wi-thin the particles, the dextrose is isomerized to a fructose rich syrup. This fructose rich syrup migrates from the part-icle and the particles are replenished with fresh dextrose.
Thus, similar to exchange of fluids in living organisms, he-terogenity between the external syrup phase and the internal particle phase exists. Certain dry immobilized isomerase preparations, such as those prepared in accordance with U.K.
patent specification No. 1,444,539, have been found to con-tain latent acidic substances (e.g., glutaraldehyde crossli-nked type of immobilized isomerases). These acidic substances are apparently initially held tightly within the structure of the dry immobilized particle. When used in an isomerization process, these acids are entrapped within the particles. The entrapped acids can create an excessively low localized pH
and cause inactivation of the isomerase.

',' ,:
~. ~ :: . . -10- ', :, Adequate processing precautlons should be taken to prevent these acids from deactivating the isomerase. This problem can be conveniently alleviated by hydrating and neutralizing the acid-containing dried immobilized particles with a non-deactivating base (e.g., sodiu~ hydroxide, bicarbonate and carbonate of sodium, etc.),prior to commencemen1: of the isomerization reaction (e.g., immediately after loading), or by adding an isomerase compatible, water-soluble base or buffer to the feed syrup in an amount sufficient to ensure maintenance of the pH between about 7.0-7.5.

The isomerization pU may be maintained within the pH 7.0-7.5 range by adding to the feed syrup or reactor a non-deactiva~ing bass(e.g., sodium hydroxide, bicarbonate, carbonate, etc.) or conventional buffers which will not desctlvate the isomerase ~e.g., sodium sulfite, sodium blsulfites, carbonates, etc.). Non-degradative reducing agent~ conventionally added I by the corn syrup manufacturer to prevent oxidative and formatlon of color } 15 bodies may be incorporated into the feed syrup such as sodium bisulfite.
. .
Although the process is conducted in the absence of heat stabiliz-ing amounts of cobalt ions, the isomerization reaction i9 conducted in the presence of co-metal ion isomerase activator. The co-metal activity requlre-ments will often vary between different isomerases (e.g., those obtained from different microbial sources). Valence two, metal ions such as mag-nesium are known and often used in isomerization as metal co-activators.
These metal ion co-activators are normally incorporated into the isomeriza-tion media as a salt (e.g., sulfate, bisulfite, citrate, acetate of mag-nesium, etc.). When a Bacillus coagula~ derived isomerase is used in the . . .
proce6~es, the presence of magnesium ions and its concentration will affect overall fructose productlvity. Although the magnesium ion molarity can -., , ''' ~"
., '',' .: ,' '.
'' ' ~"

-11- '; ', l~BE~44Z
;

range between about O.OOlM to about O.OlM for a Bacillus derived isomerase -reaction, improved productivity is obtained when the isomerization media contains at least about 0.002M magnesium ion. Above the O.OlM level, the magnesium ion requirements of the isomerase are met and any further amounts ~ -thereof merely reflects in increased processing costs (e.g., magnesium salt costs and additional burden upon refining systems to remove the ions, etc.).
A magnesium ion content between about 0.003M to aboutO.OlOM has been found to be particularly effective for enhancing total isomerase productivity.
.. . .

The bed is used until the bed activity decreases to less than 20%
and preferably less than 15% of its maximum operational activity rating.
Typically, fresh immobilized isomerase will have an assay activity (per the Ptandard assay condition mentioned above) of at least 400 International Glucose Isomerase Units ~i.e., IGIU) per 8ram and will typlcall~ be used in the continuous process until it has an activity rating of less than 80 IGIUtgm.
The most appropriate syrup flow rates through the column or bed will depend upon the desired degree of fructose conversion, bed isomerase activity, bed flow and pressure drop characteristics. In order to maintain a constant fruc-tose yield (e.g., about 45% fructose and 55% dextrose), the feed syrup flow rate is proportionally regulated so as to coincide with the isomerase activity of the bed. For example, a bed loaded with 30 pounds of isomerase per cubic foo~ of an isomerase activity of 500 units/gram (i.e., 6,810,000 IGIU/ft.3?
! and a desired output of about 45% fructose, will typically be initially operated at a flow rate of about O.l gallon per minute (gpm)/cubic foot of bed tft.3). However, when the isomerase activity decreases to about 50 IGIU/
gram (i.e., 681,000 IGIU/ft.3) a slower syrup flow rate of O.Ol gpm/ft.3 is needed to provide an equivalent fructose yield. Typical terminal syrup ) ~ ;, :.

: ' '''''.. ,`:;

,~, .

4z flow rates are greater than 0.005 gpm/ft.3 and most usually above about 0.0l gpm/ft.3. It is deslrable to use ~he highest possible flow rates.
Total bed activity, however, limits the rate at which the syrup may be passed through the column to achieve the desired fructose level. Below the 40 IGIU/gm. level, the flow rates are so slow that the process becomes ~ -uneconomical. For most column operations, the syrup flow rate will range between about 0.01 to about 0.2 gpm/ft.3.

The immobilized isomerases under the isomerization conditLons of this invention typically have an initial low activity (e.g., less than 200 IGIU/gm.) and produce a relatively low amount of fructose during the inltial 10-15 hours. In contrast, a bed operated at pH 8.5 characteris-tlcally ha~ ~ub~tantlally hlgher activity and yields more fructose.
i Atyplcally, however, the immobilized lsomerase under the present process conditions evinces a substantial actlvity increase after about 200-400 hours of use (e.g., 235 IGIU/gram or higher) while one operated at a higher pH and temperature will progressively decrease in activity. After about 400 hours, the immobilized isomerase gradually decreases in activity ,: , .
until the bed becomes spent (typically greater than 3r500 hours and prefer-ably after at least 4,000 hours of continued use) as opposed to the rapid decrease experienced ~ith immobilized beds operated at pH's above or below thi~ level. The overall net effect of the processing conditions employed in this invention is to maintain a much higher level of fructose productivity over a more prolonged period of time. ~ ~
,.~.

..
, `
:
. - .

' . .. ' :

. . . .

1~8~4~2 ~

The degree of permanent isomerase deactivation arising out of column operations conducted outside the prescribed pH 7.0-7.5 range is directly re-lated to the extent the pH deviates ~herefrom and its exposure time. On an equivalent time basis, column operation at a pH above 8.5 or below 6.0 causes more extensive permanent deactivation and reduced productivity than operations conducted within pH 7.8-8.0 or 6.5-7.0 ranges. Operation at either a pH of about 6.0 to 8.5 for a short period of time (e.g., 24 hours), generally results in a lesser degree of inactivation than prolonged operation (e.g., 300 hours) at pH's of about 6.5 or 8Ø

Due to occluded or tightly boundacidic substances contained within the immobilized isomerase (e.g., glutaraldehyde, cross-linked immobilized isomera~e), it 1~ dlfficult to immediately achieve the de~ired pH 7.0-7.5 ef~luent level. Hydrnting and neutralizing the isomerase bed ~rior to con~enc-ing the i~omerization process a~ mentioned above) accelerates the rate at which the bed will stabilize to a pH 7.0-7.5 (e.g., typically within a day). :
In contrast, the untreated beds will usually require at least two times more time for the effluent stream to stabilize to a pH 7.0-7.5. Alternatively, buffers may be used as a processing aid to maintain the pH within the pre-scribed pH 7.0-7.5 range. In the event neither buffers nor a pretreated isomerase are used and the isomerase bed co~tains a relatively high level of occluded acid, the pH of th~ influent stream may be temporarily ad~usted to a sllghtly higher pH (e.g., 7.5-8.0 and at higher flow rates) in order to compensate for the acldic bed substances and achieve the desired pH 7.0-7.5 effluent stream. ~ ~

~i 25 Is is important during the initial 2,000 hours of column operation ~ -` to maintain the isomerase bed, as reflected by effluent stream pH,~at a pH of at least 7.0 up to a pH 7.5. Enhanced isomerase productivity will be accom-plished by operating the column within this range at least 90% of the opera-` tional time, and ~ost preferably more than 95% of the operation time at pH
~ 7.0-7.5. As a general rule, the isomerase bed under the process conditions ~(~8~44Z
herein typically retain more than 40% (often 50% or more) of its original 24 hour activity rating after 2,000 hours of continued use. During the latter life cycle of the isomerase bed (e.g., after 3,000 hours), maintaining the pH within 7.0-7.5 will enhance productivity to a lesser degree. Its effect upon productivity is less ~ecause of the substantially reduced level of bed potency or activity. In the preferred embodiments of the invention, the p~ is maintained within the 7.0-7.5 range for at least 90% of its opera- -tlonal life with no more than 5% of the total operational time being above ~ -pH 8.0 or below 6.5. Substantial and permanent isomerase inactivation occurs when the process is conducted for 100 hours or more above pH 8.0 or below pH 6.5. If the pH deviates from the pH 7.0-7.5 range, isomerase activity is partially restorable by readjusting the pH to the 7.0-7.5 level. This, however, doesn't correct the permanent isomerase degradation and productivity "
108~ which has been cau~ed by operating outside the 7.0~7.5 pH range.

Temperature has a similar effect upon fructose productivity. Ele-vated temperatures temporarily increase bed activity, but decrease its over-all productlvity. The equation: lne Y =14.484 - (0.10247)(X), wherein Y is `
fructose productivity and X ls the isomerization temperature (in C.), closely approximates the affect of temperature upon productivity in the present process.
Short intermlttent exposure to elevated temperature may occur, but is desir-ably avoided. Comparative to a 65C. process, a 60C. or lower operational temperature initlally produces less fructose. Equlvale~t fructose production :
(i.e., total fructose yield) at the 60C. isomerization temperature is ~ ` ;
typically not accomplished until after 140~ hours of continuous opera-tion, at which time, the 60C. fructose productivity approaches or begins to exceed the 65C. productivity. The isomerase bed remains operationally viable at 60C. for a longer tlme period than at the 65C. level (e.g., 4,050 hours vs. 1,680 hours). Improved productivity is accomplished by maintaining the column reactor below 61C. for at least 90% of its operational life, and pre-erably at about 60C., or less, for at least 95% o its operational period.
,, .;`

1 15 ~

l(t!3~4~ ~

If desired, the column reactor temperature may be increased to 65C. or higher, -when the bed approaches the state of exhaust~on. Operational temperatures below 55~C. do not adversely affect the productivity of the isomerase bed but the reaction rate is substantially slower. When the operational temper-ature decreases below the 55C. level, the high solids syrups used herein be-come more viscous and less mobile. Microbial infestation of the bed is al~qo difficult to prevent at the reduced operat~onal temp0ratùres. From an over-all -processing viewpoint, it is advantageous to conduct the isomerization reaction between the 57C. to 61C. range.
. '. ' ':
In the continuous productlon of high fructose conversion syrups, conventional batch assay tests and conventional techniques employed to ascer-tain optimum operational conditions are misleading and frequent:Ly inapplicable to the over-all conditions needed to achieve maxlmum productivity. These conventional assay tests ~e.g., IGIU assay tests) are useful in determining the initial isomerase potency or activity rating and its suitability for use ' in a continuous high fructose syrup operation. Thereafter the most meaningful '~ guideline is the total amount of fructose actually produced by a given amount of enzyme. In a continuous high fructose syrup process, the amount of fructose produced by the isomerase bed at any ti=e during its operation ~as well as the total amount of fructosewhich has been produced) may easily be determined by monitoring the effluent atream. Periodic monitoring of the effluent stream ensbles one to ascertain when the isomerase has reached its maximum level of production. Thereafter, the bed efficency, during any stage of the process, - ~
I ~ can be determined by comparing the monitored level of fructose production with . .
~,~ 25 it9 maximum level of fructose production. In the present process, the bed is advantageously used until the amount of fructose produced by the bed decreases to a vslue of less than 15X of its maximum fructose output (i.e., its highest j ~ monitored fructose production or activity level). Preferably the reactor bed I I is replenlshed or replaced with fresh isomerase when the bed activity de-creases to within about 10% to about 15% of its maximum output level.

~l .
, -16-j .
.. . . . . . . . . . . ... . . .
.: . :, - , .. . . ~, . .. .. . . .. . . .

z In general, the isomerization conditions of this in-vention can extend the useful and productive life of an immo-bilized isomerase by at least 2,000 hours. By Inaintaining the pH within the pH 7.0-7.5 and below 61Cn ~ the present pro-cess typically permits the fructose syrup manufacturer to con-tinuously operate the column reactor and achieve adequate fru-ctose interconversion for more than 3,000 hours and most typ-ically more than 4,000 hours. The anticipated useful bed life in a pH 8.5 and 65C. process is less than :L,000 hours. The ~
present process produces a higher quality fructose syrup. ;
Syrup degradation arising from the isomerization process is nominal with minimum off-flavor and off-color development.
The present syrups can be placed in a useable form without necessitating extensive carbon and ion exchange treatment.

The following examples are merely illustrative and ~ ;

should not be construed as limiting the scope of the invention.
: '..

This example illustrates improved dextrose produc-tivity employing an immobilized glucose isomerase identified as "Novo SP-113E" sold by Novo Enzyme Corporation, 1830 Mamo-roneck Avenue, Mamoroneck, New Jersey 10543, in a single pass column type reactor under the isomerization conditions of this invention. The enzyme was derived from Bacillus coagulan im-mobilized by glutaraldehyde crosslinking under U.K. patent specification No. 1,444,539. ;

The dry immobilized isomerase was ~ hydrated and neutralized at pH 7.5 and 50C. for one hour, by slurrying 20 grams dry isomerase in 150 ml. refined 95-96% dextrose syrup (at 50% dry solids) which contained 0.005M MgSO4.7H2O adjus-30 ted to pH 7.5 (sodium hydroxide). The sodium hydroxide tre-ated and highly swollen isomerase slurry was then transferred to a water jacketed, column reactor (height 30 cm and 2.5 cm diameter.
..
~~`3 - 17 -8~4~

In this example, a 50% dry weight solids, 97.5% D.E., 96%
dextrose, 2.5% disaccharide, .5% trisaccharide and 1.5% polysaccharide tD.P.

4 or higher) feed syrup was used. The syrup was refined by treating it at -pH 4.0-4.5 with .5-1.0% powdered, activated carbon, removing the insolubles therefrom by filtration, followed by ion exchanging with a strong cation resin (Rohm & Haas Amberlite 252), weak base anion (Diamond Shamrock Duolite ES-561), strong cation (Amberlite 252) and then a weak base anion (Duolite ES-561) connected in series. The refined syrup was adjusted to pH 7.5 (sodium hydr-oxide) with 0.005M MgSO4.7H2O (as a metal ion activator), and 0.8 grams methyl paraben, 0.2 grams propyl paraben and 1.97 grams sodium benzoate (as preservatives) being added thereto (grams/syrup liter basis). The refined dextrose feed syrup (at 60C.) was continuously fed to the column reactor at a ~low rate ranging between 0.6-2.4 ml. feed syrup/minute~.
., . , ' .

¦ The pH of the influent or dextrose eed syrup stream was maintained at about 7.5 and the efluent pH stream was periodicslly monitored. During l, the inltial stages of the isomerization reaction (at about 24 hours), the i effluent stream pH gradually increased from 6.0 to the desired 7.0-7.5 range.
~ Thereafter the effluent stream pH essentially remained within the 7.0-7.5 ¦ range excepting monitored pH's of 6.9 at 336 hours and 689 hours and 8.0 at 2177 hours and 8.5 at 2201 hours. Illustratlve monitored flow rates, percent o total dextrose isomerized to fructose, bed isomerase activity, total cumula-tive ructose yields or productivity and efluent pH values ~approximately 24 ~ 7 hours basis) are tabulated in Table 1.

~ . . ..
~; For comparative purposes a column reactor was then operated under ~25 essentially the same conditions including the 60C. feed syrup, except a pH
8.5 dextrose feed syrup was used. The results of this comparative test are -. .
¦ ~reported in Table 2. ~ -f: : :
,~ . . .

1 -18- ~ ~
~! , ' ':

"
4Z ::

. ~ .

~;...................................................................... ' .
,:, .
-~ i ~1 ul u~ u~ u~ u~ u~ u~ u~ u~ u~ u~ u~ u~ u~ u~ u~ u~ u~ u~ u~ u~
. ~ ~ P l ~ r~ r~ r~ ~ r~ ~ r~ r~
: .'~ ~ ' , ~, , .
'` 1 .
~ ml . . . ~ ~ ~ N ~ ~7 N 0~ ~) ~ N ~ u~ O tt7 U~ U~ U~ ~
~,~ P~ D r~ ~ r~ r~ ~ r~ r~ r~

p ~ x ~D ~D U~ o o ~ r~ o~ U~ u~ N r~ O _I N r u~
. ~ Fl ~ ~ ) ~ ~ O O ~ ~ i ~ N _i X ~ i `D N 11 C P N ~ ~ ~1 ~D O r-l U`, u~ D O u~ D O U~l _I Ul :1 P I_t ~~ N ~ ~ U~ D r~. r~ oi O C~ ~ U~
", P E~ ,_1 _1 ~_1 _1 ,1 ,1 ,1 , ....
,,j .
:
:

.~ I ~ oo ~I ~ ~ O~ O O oo e~ ~ X 00 ~O
~ ~ I r~`J r l 1~~ I r i r~ r l ~I r~l r~l , ~ P ~ ~
, 3 ,;j ~ .......
':1 ~ .....
O r~l O N `D _I O CO C~ ~ ~ X N ~1 ~ O ~D U~ U~ r~
~t r~ 1 U~ ~ N N ~ ~ D O _~ r~l N ::.: .-, ,'' : 1 P~ . ', :' '.
'i ' . , ,. ,:':
~ ~D ~ 00 ~t ~ ~ ~ O ~ ~ O O O X C`~ CO O O ~' " ' . .
~ æ ~ ; ~ ~ ~ ~ ~ o U; N ~ U ~ ~ ~
P4 r~ r~ ", "~
: j ' ~ : ' U ~ ~ t~ I ~ r r l ~7 ~ ~--I ~1 Ul ~ ~ ~ ~ ' :~1 : 1-1 ~1 ~1 ~1 _I O O r~ l r ~I r~l r~l rl ~I r~ I r~l ~1 ~1 :
;~! P$ ~ t ~ : :.:
,~ ~i _i ri _I ri r~ i r~ 1 rl ~1 _I ~1 ~1 ~ ~1 , ~:
: ; , . :'~
i, ~ : . ',, :
c~ i3 ~ o~ ~ o ~ oo ~ ~ ~ ~ N ~ ~I X e~ 1~ u~ O r~
. : ~ u~ o o u~ o~ oo ~ oo ~ ~1 o a~ ~ ~ u~ ~D r l O~ ~D U~ ~ ~ ,, ,~., r ~ N N r l r l r 1 r~ _I r l r~ r l r~l r Jf r~l ~1 i : . . , .:
. .
:'' ." : ^ : , .
u~ r~ ~ ~ r~ ~ ~ x ~ tn ~D ~ 1~ r~ _I Y~ _I r~ r~
~I N ~ ~D ~ ~1 ~D ~ X O ~ ~ ~ ~ N ~ ~ ~) U~ U~
,,~ ~ u~ ul ~ r~ x ~ O~ ~
: ", :

19- : . ' :

~(~884~Z ~ ~

. .:

'; , : . .

,. . .

00 0 e~ O ~ ~ C~ ~ ~ ~ O~ O
,~ O ~ O O~ ~ ~ O~ ~ ~ ~ u~ ~ ~D CO 00 ~ ~ ~

.. . .

.:~ . U cn X ~ ~ ~ I~ ~ ~1 ~ ~ X ~ ~
~P
.iij ~

~i ~i ~ ~ ~ o~ co cr~ o c`i ~ ~ ~

o a~ i oo oo ~ o ~ oo u~ ~ O ~ ~D ~ ~ : ~ " ,., '. ,.

" ~ S~ O r~ X ~ O

S~ ~ ~ ~ ~t ~ O O ~ ~ ~ ~ ~O ~D ~t ~ O ~ ~O ~ : ' . -:

~ ~ o u~ o o ~ ~
' ~ ~ ~ ~ ~~ ~ ~ ~ I~ a~ o o ~ ~ "

` : .

20- `

., i .. ~ . ; . . ~ : . . . .

l~B~44Z ~

F4, 1 .
, ,, P ~ ~ o~ ~ ~ ~1 ~D ~ ~ ~ ~ ~ `D ~ :
o P C~l ~ ~ ~ X ~o ~ o~ ~ ~ ~ U~ ,~ ,,, ,, ~, >~ ~ , ::
l O U~ "
-l 1`. O~ ~ ~O ~ Itl b'~ O O~ 00 1~ ~ O~
. ' ~ ~
~ ~P . ~;.
~ ~. ' X
o~ o r~ i o ~ x ,i o o tr; ~ , ~ ..

: .

~' ' ., ~ ~ ~ ~ ~ ~ O ~ l O ' .
~,7 ~ ~ I O O ~1 ~ ~1 ~I

:1 :

:~, o~ ~ ~ ~ D . "

:l : ^

: :

.j ^ ~.

.

... . .. . . . . . . .. ... . ..... . ...

108~4~ ~ ~
. . - .
As shown in Table 1, isomerase activity and fructose yields for the 7.0-7.5 isomerization reaction was initially low but gradually increased to the optimum (monitored) isomerase activity of 2491 units/gram after about 10 days operation. The isomerase activity for the comparative pH 8.5 reaction 1~ ;
was initially significantly higher and reached an optimum monitored activity after about one day. Although inltial productivity for the pH 7.0-7.5 process was lower than the pH 8.5 process its productivity exceeded the pH 8.5 process after about 5 days operation. After about 22.7 days operation the cumulative fructose productivity for the pH 7.0-7.5 reaction was 1069.7 versus 676.2 for the pH 8.5 syrup (Table 2).

As illustrated by the Table 1 data, isomerase activity and fructose productlvity decreased when the e~fluent stream dropped below the pH 7,0 level but lncreased upon its subsequent restoration to the pH 7.0-7.5 (e.g., co~pare the 305, 336 and 353 hour data). The Table 1 data also shows 1 15 that an effluent pH of 8.0-8.5 wlll significantly reduce both the isomerase ~-¦ activity and fructose yields with higher fructose production levels occurring ~,~ upon restoration to the pH 7.0-7.5 range (e.g., see 2153-2223 hour data).
,' These excessive pH levels are undesirable even though it occurred late in the continuous process and they adversely affected over-all fructose pro-ductivit~ of the bed. ~

Remaining or residual isomerase activity (based upon 17 hour activity ~ ;
at 100%)for the column reactor operated continuously with a pH 7.5 feed ~: ......
syrup and 3,000 hours was in excess of 30% and greater than 28% after 3,200 ~
, ¦~ hours of con~inuous operation. On the basie of the isomerase beds maximum :
¦ 25 monitored activity level (236 hours), the 25% activity level or 75% de-¦ activation dld not occur untll about 2,500 hours. About 2,978 hours of ~ ~ operational use were required for the bed to decrease to the 20% activity -¦ ~ level with the calculated 15%, 12.5% and 10% activity levels respectively . ....
~ ~ occurring at about 3,555 hours, 3,920 hours and 4,368 hours. Comparatively, ~-!

~ 22~

49~2 :- .
the pH 8.5 feed reactor will yield a calculated 20% activity level at about 990 hours and for 15% activity at 1,175 hours, a 12.5% at 1,292 hours and a 10% activity level after about 1,436 hours of operation. Total calculated productivity to the 10% activity rating (based on its maximum column activity rating) for the pH 7.5 feed is 3,~73 and 1,017 for the pH 8.5 process. Ave-rage isomerase half-life for Table 1 run was 1,389.64 hours versus 445.33 hours for the pH 8.5 Table 2 run.

The fructose-rich effluent stream of the Table 1 run was clearer, :
contained less contaminants and requlres less refining to place it in a commercially acceptable form than Table 2 run effluent stream.
. ' ~ ~. .
.` ''~: .
In Tables 1 and 2 (from left to right) the continuous operational time is reported ln the column headed (time/hrs.), the flow rate is cubic centimeter3 per minute8 under the heading"f~ow cc/min." the '~eas. R.I."
represents the refractive index of the effluent syrup (at 45C.), the '~eas.
R0TA" is the optical rotation of the effluent fructose syrup (@ 20 cm cell ,l path and 25C.), the activity U/G Enzyme is the calculated activity rating ;l of the isomerase on unit per gram of isomerase basis, the productivity ! column reports the total cumulative amount of syrup dry substance produced .. . . .
by the column upon the basis of isomerization to a 42% fructose synlp (dry ;~
,~ 20 solids basis), EFF. pH is the effluent stream pH with the feed pH being :
reported ln the extreme right column.

The reported percent fructose values for the effluent stream were :j s~ determined from the "Meas. R.I." and "Meas. R0TA" values by the following -computations:
Weight % dry substance or "%d.s." = (487.74)(Meas.R.I.) - 639.72 Speciflc Rotation or "S~.R." = ( (2 4)(% d % Fructose or %F = 38.7 - (0.63)(S.R.) 1 , :', :.

1 . ' .:
~ . .

.
, , . -:

.. : : - . ~ . , ; . .. . ,. .. . . : .: . : , . . : ,. , . , . . : . :

~108~3~42 : ~

The "activity U/G Enzyme" represents the isomerase activity of the bed (in fructose activity units/gram isomerase) at the specified time intervals which -were computed as follows:
Isomerase equilibrium constant or E.C. =
(0.513)(% monosaccharide dry substance in feed syrup) Activity U/G Enz. =
(% d.s.)(60)(cc/min.)(10)(1.2)(E.C.) X 1 r E.C.
gms. enzyme ne ~ C.-%F/10~
wherein lne is the natural logarithm, 60 (cc/min.) represents feed syrup flow rate in cubic centimeters per hour and gms enzyme is the total grams of ' isomerase in the column.

; Prior to column loading the initial activity rating for the "Sweetzyme E" immobilized isomerase employed ln this Example (per the ~i 15 standard assay methodology mentioned on page 8, lines 16-20 above) was 600 Glucose Isomerase Units/gram of isomerase.2 In a larger column operation, maintenance of the pH within the -;
7.0-7.5 range i5 easier (e.g., internal pH sensory devices) than in a smaller column operation. Similarly in a large column operation a constant percent fructose syrup content can be more easily controlled by a effluent stream fructose analyzer so as to permit manual or mechanical regula~ion of the feed syrup flow rate to yield a constant fructose level in the effluent fructose stream.

; The dry immobilized glucose isomerase employed herein may be pre-treated for 30 minutes at ambient temperatures (e.g., 23C.) and pH 7.5 with dextrose feed which contains cobalt ion (e.g., O.OOLM cobaltous nitrate) 1 - Total % fructose and dextrose (dry substance weight basis) in feed syrup which was 96%.
2 - In micromoles/min. vs "activity U/G Enzyme" in mg./hour . ~ . .
-.. - . :

and 0.02M magnesium sulfite, for purposes of hydrating and assuring the isomerase contains its full complement loading of cobaltous ion prior to commencement of the isomerization run. ; -This example illustrates high fructose productivity without adding any cobaltous ions to the feed syrup. Although ;
not illustrated herein, cobaltous ions in amounts such as em- ~
ployed in conventional isomerization operations (e.g., O.OOlM ~-Co+ ) adversely affect isomerase productivity. If desired, low cobaltous ion concentrations (e.g., less than 0.0005M), and advantageously at concentrations less than 0.00025M Co++
(most preferably less than 0.00005M) may be continuously added ~ ~
to the feed syrup stream. Alternatively, somewhat higher co- ;`

`'! baltous ion concentrations (e.g., less than about O.OOlM) may be intermittently added to the feed syrup.
. , ;~',: .
EXAMPLE II

Employing the refined dextrose syrup and column re~
actor system of Example I, two different feed syrups were con- ~
tinuously fed to a glucose isomerization column reactors. The `
isomerase (Novo SP-113B) was derived from Bacillus coagulan, I immobilized via glutaraldehyde cross-linking and spray-dried per the disclosure of U.K. patent specification No. 1,444,539.
Screen analysis for the spray-dried immobilized isomerase was 69% on ~20 sieve. In one run, a dextrose feed syrup contain--~ ing 0.008M sodium bisulfite (as a buffer~ and O.OO9M epsom -3~ salt (MgSO4.7H2O - metal ion activator) at a pH 7.5 and 60C.
was continuously introduced to the column reactor at flow rates ranging between 1.62 to 6.18 cc/minute. The feed syrup in the comparative run was the same, excepting the feed syrup was adjusted to pH 8.5 without added sodium bisulfite buffer and maintained at a flow rate ranging from 0.80 to 4.54 cc/
minute.

~.., ., ,.., .. ~., i B _ 25 -Similar to Example I, the pH 7.5 run was significantly more pro-ductive than the pH 8.5 run. After 1,891 hours continuous usage, the pH 7.5 run retained 28~ of i~s maximum isomerase bed activity with a productivity of 3,011. The pH 8.5 run had deteriorated to a 21% activity rating after 833 hours with a productivity of only 838.
, EXAMPLE III
The Example II pH 7.5 run was repeated at different isomerization temperatures ranging between 55 to 65C. The initial isomerase activity at 65C. was appreclably higher than at 60C. (1.5 times greater), but upon pro-longed usage its productivity and activity was substantially below the Example II pH 7.5 run. Calculated productivities at different operational temperature~ and continuously conducted until the isomerase bed activity had decreasecl to 10% o~ their re8pective inltlal activity ratings were 6,994 at 55C., 6,295 at 56C., 5,670 at 57C., 5,110 at 58C., 4,608 at 5gC., 4,158 at 60C., 3,754 at 61C., 3J392 at 62C., 3,066 at 63C., Z,773 at 64C., . .
and 2,510 at 65C. Column operation at temperatures from 55C. to 60C.
result in a total fructose production yield increase ranging from 179% to 66%
over those obtained at 65C.

, ,: .
Since many embodiments of this invention may be made and since many changes may be made in the embodiments described, the foregoing i9 interpreted as illustratlve and the lnvention is defined by the claims appended hereafter.
, ' ', '' ~ ' ,-~ .

' , .

Claims (11)

The embodiments of the invention in which an exclu-sive property or privilege is claimed are defined as follows:

1. A process for increasing fructose yields in a glucose isomerization process employing a fixed bed in which the isomerase is characterized as being obtained from the Bacillus genera and exhibiting an enhanced rate of isomeriz-ing dextrose to fructose when the glucose isomerization re-action is conducted: (a) in the presence of Co++ ions and (b) at a temperature greater than about 60°C., said process com-prising:
(a) providing a refined monosaccharide feed syrup which is essentially free from Co++ ions and containing on a total dry solids weight basis at least 90% monosaccharide;
(b) isomerizing the feed syrup to fructose by pass-ing the syrup through a bed of immobilized glucose isomerase at an isomerization temperature ranging from about 55°C. to 61°C. and an isomerization pH between 7.0-7.5;
(c) recovering the isomerized syrup while replen-ishing the bed with a compensatory amount of feed syrup; and (d) continuing the feed syrup isomerization in said bed, while substantially maintaining the isomerization temperature between about 55°C. to 61°C. and the isomeriza-tion pH from 7.0-7.5, until the isomerase activity of said bed is less than 20% of its optimum activity level and there-by increasing the amount of fructose recovered from said bed.

2. The process according to claim 1 wherein the monosaccharide isomerization in said bed is continued for more than 3,000 hours.

3. The process according to claim 2 wherein for more than 90% of the operational time the isomerization is conducted at a pH from 7.0 to 7.5, a temperature from about 58°C. to 61°C. and a cobaltous ion concentration of the feed syrup less than 0.00005M.

4. The process according to claim 1 wherein the feed syrup has a dextrose content of at least 95% (dry substance basis) and a dry solids content ranging from about 50 to about 55% by weight.

5. The process according to claim 4 wherein the dextrose feed syrup isomerization is conducted in the presence of an isomerase derived from Bacillus coagulen.

6. The process according to claim 1 wherein the monosaccharide content of the feed syrup is at least 95% (dry substance weight basis) and the feed syrup is isomerized in said bed for more than 3,500 hours.

7. The process according to claim 6 wherein the feed syrup con-tains at least 0.002M magnesium ion.

8. The process according to claim 1 wherein the isomerization pH
is maintained within the 7.0 to 7.5 range at least 90% of the operational time with no more than 5% of the total operational time being conducted at a pH above 8.0 or below a pH 6.5.

9. The process according to claim 8 wherein the feed syrup is characterized as having a monosaccharide content on a dry substance weight basis of at least 92%, a magnesium ion content from about 0.003M to about 0.01M, a dry solids weight content from above 45% to about 55%, and a cobaltous ion concentration of less than 0.00005M.

10. The process according to claim 9 wherein the feed syrup isomerization is continued for at least 3,000 hours.

11. The process according to claim 10 wherein the isomerase is derived from Bacillus coagulan and the feed syrup isomerization is continued until the isomerase bed activity reduces to less than 15% of the optimum isomerase activity of said bed.